1. Folkes v. Chadd, 3 Doug. 157, 99 Eng. Rep. 589 (1782), an eighteenth-century English case often cited as the first in which a court allowed expert opinion testimony, involved engineering experts who provided conflicting explanations for what had caused a harbor to become clogged with silt. See Tal Golan, Revisiting the History of Scientific Expert Testimony, 73 Brook. L. Rev. 879, 886–904 (2008).

2. Based on a Westlaw search of engineering expert cases decided during the 18 months from January 2021 through June 2022 (discussed in detail below, in the section titled “Analysis of the Case Law Based on an Understanding of Engineering and How It Differs from Science”), federal courts currently decide about 240 cases a year involving engineering experts, of which only 134 involve the kind of complex engineering testimony on which this reference guide focuses. In just over 20% of the 240 cases, the experts were accident investigators, an area with which courts have become quite familiar and comfortable after a history going back more than a hundred years. See, e.g., Grand Trunk Ry. Co. of Canada v. Ives, 144 U.S. 408, 412 (1892) (expert testified that the distance traveled by a train after colliding with a buggy indicated “it must have been going at the rate of 25 or 30 miles an hour”); Shugart v. Atlanta, K. & N. Ry., 133 F. 505, 507–08 (6th Cir. 1904) (allowing conflicting expert testimony about whether condition of railroad tracks had caused a derailment). Fifteen percent of the cases involved insurance coverage issues, and about 7% involved matters like employment claims or criminal law issues that don’t really implicate complicated engineering questions. Again, this reference guide focuses on the more complex engineering issues that arise in the remaining 58% of the cases, the majority of which involved products liability claims.

3. Henry Petroski, Reference Guide on Engineering Practice and Methods, in Reference Manual on Scientific Evidence 579–624 (2d ed. 2000).

4. Channing R. Robertson, John E. Moalli, & David L. Black, Reference Guide on Engineering, in Reference Manual on Scientific Evidence 897–959 (3d ed. 2011).

5. 509 U.S. 579 (1993).

6. 526 U.S. 137 (1999).

7. Daubert, 509 U.S. at 593–95; Liesa L. Richter & Daniel J. Capra, The Admissibility of Expert Testimony, in this manual at 5–13 (hereinafter Richter & Capra).

consideration of a multitude of hypotheses, for those that are incorrect will eventually be shown to be so, and that in itself is an advance.”⁸

Daubert also provided a “five-factor, nondispositive, nonexclusive, ‘flexible’ test” for trial courts to use when determining the validity of scientific evidence, as required by Rule 702.⁹ The five factors, generally referred to as the “Daubert factors,” are:

1. whether the technique or theory can be or has been tested;
2. whether the theory or technique has been subject to peer review and publication;
3. the known or potential rate of error;
4. the existence and maintenance of standards and controls; and
5. the degree to which the theory or technique has been generally accepted in the scientific community.

In the immediate aftermath of Daubert, there was uncertainty about whether its gatekeeping mandate applied to all expert testimony or just to scientific experts, and if it did apply more broadly, whether the five factors should be used for nonscientific experts.¹⁰ Kumho Tire resolved the uncertainty about broader applicability of Daubert by extending judicial gatekeeping to all expert evidence, but provided no analogous general guidance or list of factors to consider when evaluating the engineering expert testimony (from a tire failure expert) at issue in that case. Instead, the Court’s decision focused on the specific evidence the trial court had excluded and concluded the trial court had not abused its discretion. The Court’s reluctance to delve into general guidance or factors for fields of expertise other than science reflects the fact that except for science (the field of human endeavor focused on expanding our knowledge of the natural world), there exists no extensive literature on the philosophical underpinnings for the knowledge experts may develop or acquire. For engineering, however, there is a wealth of practical guidance that forms the basis for this reference guide, which addresses three fundamental questions: (1) What is engineering, and how does it differ from science? (2) Who is an engineer? and (3) How do engineers think? In particular, how do they create, test, and evaluate designs? As explained in the

8. Daubert, 509 U.S. at 597. See also Michael Weisberg & Anastasia Thanukos, How Science Works, in this manual, which makes clear that textbook and media versions of the scientific method (hypothesis, followed by experimental testing, followed by publication, followed by acceptance and consensus in the scientific community) is far too linear and simple, and does not reflect how science actually works in practice. How Science Works, at 50–56.

9. For more on Daubert, Rule 702, and admissibility, see Richter & Capra, supra note 7.

10. Compare Watkins v. Telsmith, Inc., 121 F.3d 984, 991 (5th Cir. 1997) (“the nonexclusive list of factors relevant under Daubert to assessing scientific methodology . . . are also relevant to assessing other types of expert evidence”) with McKendall v. Crown Control Corp., 122 F.3d 803, 806 (Daubert factors “are relevant only to testimony bearing on ‘scientific’ knowledge”).

following section, the creation and analysis of designs is the central feature of engineering practice, and it is the primary focus of this reference guide.

What Is Engineering and How Does It Differ from Science?

Engineers do often use or apply scientific knowledge, but that does not mean engineering is in any way subsidiary to science. Indeed, the practice of engineering pre-dates science by thousands of years. There is, however, an interplay between the fields, which means judges sometimes may have to decide whether to apply some or all of the Daubert scientific evidence criteria or different engineering criteria when determining the admissibility of an engineer’s testimony.¹¹ Thus, the necessary starting point for this reference guide is a discussion of how the two fields differ and how they relate to each other. What, then, distinguishes engineering from other fields, especially science? Theodore von Kármán, an eminent aerospace engineer (also a mathematician and physicist) once explained that “scientists study the world as it is, engineers create the world that never has been.”¹²

More precisely, engineering and science differ in three important ways:

1. Historically, many engineering products and technologies have preceded scientific understanding,¹³ and while this remains true today, the advent of modern science often allows engineers to use scientific discoveries as part of their tool kit when they create new designs.¹⁴ In no event, however,

11. For example, if called upon as an expert witness to determine the cause of an accident or the failure of a product, an engineer might take various approaches. This reference guide is most applicable when an engineering expert bases such a determination on the analysis of a design. See, e.g., Baugh v. Cuprum S.A. de C.V., 845 F.3d 838 (7th Cir. 2017) (in which a mechanical engineer relied on calculations based on centuries-old mathematical principles to determine the cause of a ladder failure). If, on the other hand, an engineering expert formulates a hypothesis that requires testing or other empirical corroboration, some of the Daubert factors used for evaluating science might be more applicable. See, e.g., Nease v. Ford Motor Co., 848 F.3d 219 (4th Cir. 2017) (in which the plaintiff’s expert hypothesized that debris in a cable housing had caused an accelerator to stick but offered no analytical explanation and no testing of his hypothesis).

12. Petroski, supra note 3, at 579. Theodore von Kármán was the National Medal of Science recipient in 1962.

13. William S. Hammack & John L. Anderson, Working in the Penumbra of Understanding, Issues in Sci. & Tech., Feb. 16, 2022, available at https://perma.cc/48H5-WUCN.

14. One good example is the use of discoveries by the Nobel Laureate Jacques Monod about how bacteria reproduce as part of the design process for wastewater treatment plants. Alexandru Braha & Ferdinand Hafner, Use of Monod Kinetics on Multi-Stage Bioreactors, 19 Water Rsch. 1217 (1985), https://doi.org/10.1016/0043-1354(85)90174-5.

15. Billings Clinic, The Scientific Method vs. Engineering Design Process, available at https://perma.cc/H6CZ-98TE.

16. Gilbert Ryle, Knowing How and Knowing That, Proceedings of the Aristotelian Society, XLVI (in Collected Essays 1929-1968, Collected Papers Vol. 2, pp. 212-25).

17. Richter & Capra, supra note 7, at 23, in this manual, quoting Advisory Committee’s note to 2023 amendment to Fed. R. Evid. 702.

18. Id.

19. The Supreme Court’s observation in Kumho Tire about engineering resting on a scientific foundation may have inadvertently created some confusion on this point. In one sense it could be

taken to mean the criteria for scientific validity and reliability should be used to evaluate all engineering testimony, but the far better interpretation is that engineers sometimes use scientific discoveries in the course of creating designs. If the scientific underpinning were not well accepted and reliable, then the engineering use of it could be called into question, but that is not fundamentally related to the engineering design process.

20. Petroski, supra note 3, at 580; Robertson et al., supra note 4, at 902. See also Hammack & Anderson, supra note 13.

21. See Hammack & Anderson, supra note 13.

22. Id.

23. See Michael Weisberg & Anastasia Thanukos, How Science Works, p. 82, in this manual.

24. Id.

25. Joseph C. Pitt, What Engineers Know, Techné 5 (3): 19 Spring 2001, available at https://perma.cc/GA4X-5Q3E.

26. National Academy of Engineering, Understanding the Educational and Career Pathways of Engineers (2018), https://doi.org/10.17226/25284.

27. The list includes sales engineering, defined in the standard occupational classification system used by the National Science Foundation (NSF) as those who “sell business goods or services, the selling of which requires a technical background equivalent to a baccalaureate degree in engineering.”

28. ABET, Criteria for Accrediting Engineering Programs, 2023–2024, available at https://perma.cc/JZ2G-QNMU.

29. IEEE Societies, available at https://perma.cc/6N8T-J3JX.

engineers similarly educated and experienced could be armed with sufficiently different perspectives that they would flat out contradict each other.”³⁰ This makes engineering testimony less about discovering or converging on “the one truth” and more about adherence to process and practice. Although neither the traditional conception of the scientific method nor any effort to precisely define the engineering process fully captures all the nuances of what happens in either field, a comparison of traditional simplified views of both science and engineering does help to highlight how the two fields differ. Table 2 shows this comparison.

Table 2. Steps of Scientific Method Versus Engineering Design Process

In addition to the points listed in Table 2, another salient difference between science and engineering is that engineered products are intended to interact with human users or as part of an engineered system. It is not enough that a phone or a bridge function properly in the controlled conditions of a laboratory or in simulation. Phones must work despite the day-to-day abuse by their users, and bridges have to withstand various weather and congestion conditions and function properly under those conditions. Engineered products must also be safe, ergonomically designed, and aesthetically pleasing, as well as functional. It is indeed the various requirements and constraints that make engineering a creative synthesis of art (idea creation), scientific knowledge of how the world works, and accumulated design experience. Science, on the other hand, is more about explanatory theories than about practical application of those theories.

30. Joseph C. Pitt, What Engineers Know, Techné 5 (3): 17 Spring 2001, available at https://perma.cc/GA4X-5Q3E.

31. Accreditation is performed by ABET (Accreditation Board for Engineering and Technology). Founded in 1932 as the Engineer’s Council for Professional Development (ECPD), it was later renamed ABET. In the United States, accreditation is a nongovernmental, peer-review process that ensures the quality of the postsecondary education that students receive. Educational institutions or programs volunteer to undergo this review periodically to determine if certain criteria are being met. ABET accreditation is assurance that a college or university program meets the quality standards established by the profession for which it prepares its students. The quality standards that programs must meet to be ABET-accredited are set by the ABET professions themselves. This is made possible by the collaborative efforts of many different professional and technical societies. These societies and their members work together through ABET to develop the standards, and they provide the professionals who evaluate the programs to make sure that the programs meet those standards.

32. Engineering & Engineering Technology by the Numbers, ASEE (2021), https://perma.cc/5YXZ-HR2Y.

33. In the twentieth century, Buckminster Fuller seamlessly combined elements of geometry (“science”), structures (“engineering”), and architecture (“art”) to conceive and develop an entirely new approach to architectural design. Architects Norman Foster and Frank Gehry seized on recent advances in computer science and engineering to develop innovative platforms for architectural design that paved the way for radical changes in structural and visual renderings. Striking examples include the Guggenheim Museum in Bilbao, Spain; the Walt Disney Concert Hall in Los Angeles; the Experience Music Project in Seattle; the London City Hall; the Beijing Airport; and the Reich-stag in Berlin. Other examples include the famous scientist Linus Pauling (Nobel Prizes in Chemistry and Peace), who was trained as an engineer, and Albert Michelson (first American to win a Nobel Prize in science), whose work designing instruments to measure the speed of light was more engineering than science.

34. See, e.g., Guay v. Sig Sauer, Inc., 610 F. Supp. 3d 423, 427–32 (D.N.H. 2022) (former firearms instructor and armorer found qualified to testify on design of pistol even though he was not an engineer and had done no work on the design or manufacturing of guns).

35. In some states, businesses generally cannot offer engineering services to the public, or cannot have a name that implies that they do so, unless they employ at least one PE. For example, New York requires that the owners of a company offering engineering services be PEs. See https://perma.cc/2WWD-9E56.

36. The Fundamentals of Engineering (FE) exam, formerly called the Engineer in Training test, is typically taken during an undergraduate student’s last year in college. People who pass this examination and graduate with an engineering degree are called engineering interns (EIs) or engineers in training (EITs). They become eligible to take the second test, called the Principles and Practice of Engineering examination, after they’ve accumulated enough experience, typically four years of doing engineering work.

exam. However, this isn’t always the case. One should make sure to determine the relevant state requirements.³⁷ To retain their licenses in most states, PEs must continually maintain and improve their skills throughout their careers. Obtaining a PE license may be critical in some fields, such as civil engineering,³⁸ but is less so in others. PEs have the authority to certify documents such as reports, drawings, or calculations by signing and sealing (or “stamping”) them, thus attesting to the accuracy and correctness of the documents and taking legal responsibility for them.

Many engineering professionals do not seek a PE license because their services are not offered directly to the public or they have no need to certify engineering documents. Whether an individual is licensed as a PE is neither sufficient nor necessary to establish her or his competency as an engineer. Furthermore, the two examinations required for licensure test only for knowledge gained and assimilated at the undergraduate level. It is therefore common for professors of engineering not to have PE licensure, even though they are the ones who teach and prepare others who take the two required examinations.

PE licensure is thus quite different from board certification for a physician or bar admission for a lawyer. Physicians cannot practice medicine without board certification, and lawyers can’t practice law if not admitted to the bar, but such strict limitations do not apply universally to engineers. While the title “engineer” is legally protected in many states (meaning that it is unlawful to use it to offer engineering services to the public without a PE license), there are numerous exceptions. People licensed as “operating engineers” are not PEs. There is an industrial exemption for “in house” engineers employed by a manufacturing company or other business not providing a service directly to the public. Employees of state or federal agencies may also call themselves engineers if that term appears in their official job title.

Courts generally have not required an expert to be a licensed PE in order to testify.³⁹ In some cases, however, PE licensure can be an important factor in determining if an expert is qualified.⁴⁰ In still other cases, courts have held that

37. See Eligibility Requirements for the PE Exam by State, available at https://perma.cc/D9RV-XBNZ.

38. Because they often are involved in public works projects, civil engineers are more likely than engineers in other fields to obtain PE licenses. This can be traced directly back to the legacy of the St. Francis dam collapse in southern California in the late 1920s. St. Francis Dam Disaster Revisited (D.B. Nunis, Jr., ed., 2002).

39. See, e.g., Emig v. Electrolux Home Prods. Inc., No. 06-CV-4791 (KMK), 2008 WL 4200988 (S.D.N.Y. Sept. 11, 2008) (expert qualified to testify even though not a PE). Cf. McRunnel v. Batco Mfg., 917 F. Supp. 2d 946 (D. Minn. 2013) (fact that expert licensed in Arkansas but not Minnesota was proffered in Minnesota did not make him unqualified).

40. See, e.g., Clena Investments, Inc. v. XL Specialty Ins. Co., 280 F.R.D. 653, 661–62 (S.D. Fla. 2012) (Citing education and experience of plaintiff’s expert and noting that like defendant’s expert, he was a professional engineer. “Presumably [defendant viewed its expert’s] professional engineering license as at least one relevant qualification . . . No less is true of [plaintiff’s expert’s]

a PE without adequate experience relevant to the issues at hand is not qualified.⁴¹

Just as an undue judicial gatekeeping focus on an expert’s degree or licensure is inappropriate, gatekeeping based only on determining whether scientific or engineering expertise is required would be a futile endeavor. Rather, the inquiry ought to focus on the extent to which the scientific hypothesis testing or the engineering design process is at issue. As an example, fire investigators are often considered engineers, and some universities even offer courses in fire protection engineering. Yet the conduct of an investigation into the origin or cause of a fire involves the kind of hypothesis testing so essential to science. The National Fire Protection Association’s Standard 921 “sets the bar for scientific-based investigation and analysis of fire and explosion incidents.”⁴² Note, however, that while the Daubert hypothesis-testing factor is important for such investigations, peer review of every investigation would not be appropriate, nor would the general acceptance of every investigation.

By way of contrast, the investigation into the 1981 collapse of a hotel skywalk in Kansas City, which killed more than a hundred people,⁴³ provides a classic example of determination of cause through analysis of a design. Without any testing or peer-reviewed publication, and based only on a careful analytical review (using accepted methods and principles), engineers from the National Bureau of Standards quickly determined that the original design was safe, but that a modification made during construction had resulted in an unintended doubling of the load on one element of the support structure.⁴⁴

comparable license and experience.”); Urda v. Valmont Indus., Inc., 561 F. Supp. 3d 632 (M.D. La. 2021) (PE license one factor considered in finding expert qualified); Klingenberg v. Vulcan Ladder USA, LLC, 936 F.3d 824 (8th Cir. 2019) (PE license one factor considered in finding expert qualified).

41. See, e.g., Dreyer v. Ryder Automotive Carrier Grp., Inc., 367 F. Supp. 2d 413, 426–27 (W.D.N.Y. 2005) (Although proffered expert held a doctorate in mechanical engineering and was a licensed professional engineer, he had no direct experience with the trailer securing system at issue in the case. When asked to explain how a designer would go about assessing the foreseeability of user risk in connection with a product, he could not answer clearly.); Taber v. Allied Waste Sys., Inc., 642 Fed. App’x 801 (10th Cir. 2016) (PE excluded because of lack of relevant experience).

42. NFPA 921, available at https://perma.cc/R9RU-MZMF. See also Elosu v. Middlefork Ranch Inc., 26 F.4th 1017, 1029 (2022) (citing NFPA 921 and admitting investigator’s testimony after noting that he “applied broadly accepted scientific principles and professional standards to conduct his analysis”).

43. In re Fed. Skywalk Cases, 680 F.2d 1175, 1177 (8th Cir. 1982).

44. Deborah R. Hensler & Mark A. Peterson, Understanding Mass Personal Injury Litigation: A Socio-Legal Analysis, Brook. L. Rev. 961, 972–74 (1993).

45. Guru Madhavan, Applied Minds: How Engineers Think (2016).

46. W. Brian Arthur, The Nature of Technology: What It Is and How It Evolves (2009).

47. Courts have addressed mathematical models in various contexts and generally admit evidence based on them so long as the inputs reflect the actual facts in a case. See, e.g., Lapsley v. Xtec,

Inc., 689 F.3d 802, 816 (7th Cir. 2012) (admitting testimony based on a “mathematical model . . . used in place of physical re-creation”); Leibfried v. Caterpillar, Inc., Case No. 20-CV-1874, 2023 WL 7284795 at *4 (E.D. Wis. Nov. 3, 2023) (expert’s manipulation of “the virtual model to produce results that approximated the post-accident condition” did not preclude admissibility of his testimony); Sommerville v. Union Carbide Corp., Civ. Action No. 2:19-cv-00878, 2024 WL 1204094 at *1 (S.D. W. Va. Mar. 20, 2024) (testimony based on model excluded because the inputs were “speculative and . . . premised on assumptions that do not accurately represent the Defendants’ operations”).

48. Arthur, supra note 46.

known technology,” and not necessarily the invention of new technology.⁴⁹ Perhaps the utilitarian notion of providing “the greatest good for the greatest number” is apt here.

Understanding Tradeoffs

Engineering has also been described as a “set of tradeoffs,” and Madhavan introduces the notion of tradeoffs as a last pillar of engineering thought. Tradeoffs are how engineers adapt to constraints. Engineers adopt design priorities and allocate resources by eliminating less important goals and adapting to the more critical constraints in order to focus on achieving the more important goals. When working on the design for a product, engineers must therefore balance various design constraints such as usefulness, usability, cost, performance, aesthetics, and time limitation, all of which make the engineering task quite challenging. A tension exists between the various specifications of the customer. Famed bicycle parts engineer Keith Bontrager is well known for his aphorism: “Strong. Light. Cheap. Pick Two.”⁵⁰ In other words, customers can choose two of three characteristics: strong, light, and cheap. He specialized in designing strong, light, and expensive components. That kind of unavoidable tradeoff leads engineers to rely on their education and training, as well as on their experience and standards to achieve a successful outcome. If one or more of these constraints is relaxed or eliminated (e.g., unlimited time), then the engineering task becomes easier and more achievable, unless the constraint is a fundamental physical law. As another example, for a given airplane design, a typical tradeoff may be balancing the demands of cost, weight, wingspan, and lavatory dimensions within the constraints of the given flight performance specifications. This selection pressure may even lead to the question of whether passengers actually like the airplane they’re flying in and enjoy the experience. If designing within constraints can be viewed as walking a tightrope, then tradeoffs are akin to a tug-of-war among what’s available, what’s possible, what’s desirable, and what the fundamental physical limits are.

Thus, Madhavan’s three essential properties⁵¹ of the engineering mindset lead to the concept of “good engineering practice” or “best engineering practice,” which guides the practitioner of the art.

Arthur makes a different point as he describes engineering design as a matter of choice. As constraints limit possibilities, the design problem becomes more complicated, which leads to a need for a larger number of and more sophisticated components in order to achieve the design goals. Somewhat counterintuitively,

49. Id.

50. Grace on Wheels, “Strong, Light, Cheap. Pick Two,” available at https://perma.cc/BE7H-DPAA (accessed July 17, 2024).

51. Madhavan, supra note 45.

as long as they do not completely eliminate feasibility, constraints actually lead to a richer set of practical possibilities. In effect, “[a]ny new version of a technology is potentially the source of a vast number of different configurations.”⁵²

Engineering Systems

System analysis has become increasingly important in engineering design. Technologies must be evaluated in ever-broadening contexts that include their performance as intended by the designer, but also how they interact with other systems. An electric car, for example, has many mechanical and electrical components as well as software modules that are tested individually, then as subsystems, and finally as a whole system as described in the next section. The hierarchy continues, however, as the car is operating along with other cars over a transportation system. The safety and risk analyses must therefore be conducted at all levels of the hierarchy, from a failure analysis of components all the way up to avoiding accidents in a dynamic traffic environment. The analysis is made more difficult as more products are engineered, monitored, and controlled using machine learning and artificial intelligence.

Systems and Systems of Systems

As mentioned earlier, engineering is a human activity, and while some animals make tools and come together in groups to achieve simple engineering feats, humans are unique in their ability to design systems made of specifically designed components, which may then be assembled into larger systems and systems of systems. A distinguishing feature of engineering is that its products and technologies are intended for use by experts and non-experts alike and must therefore be useful and safe as they interact with living and nonliving objects. Understanding and/or operating sophisticated systems or high-complexity systems, such as flying an airplane, will of course require advanced training and certification, while driving a car requires much less. In addition, engineering systems interact with other systems, leading to systems of systems where the interactions between and among systems are just as critical as the behavior of each individual system.

Consider, for example, the social goal of transporting people and goods efficiently while keeping vehicular congestion at an acceptable level. A modern transportation system is composed of vehicles, road networks, railroads, shipping lanes, air traffic lanes, policies, governance structures, and more. A vehicle is one such system, composed of many electrical and mechanical subsystems.

52. Arthur, supra note 46.

Putting all the transportation sytem’s subsystems together requires the expertise and contributions of civil engineers, mechanical engineers, electrical engineers, and computer engineers, as well as materials engineers, policy makers, psychologists, and financial and other experts.

Some courts have recognized that engineering products are a collection of components and systems, and the value of expertise in designing and analyzing such systems. Campbell v. Fawber⁵³ provides an outstanding example, and shows how engineers can address complex causation questions in the context of litigation. The plaintiff was a passenger in a sport utility vehicle (SUV) that rolled over when the driver lost control.⁵⁴ The plaintiff suffered catastrophic neck injuries that left her quadriplegic.⁵⁵ She sued both the driver and the manufacturer of the SUV, the latter on the theory that the vehicle was not crashworthy because the roof crushed too easily.⁵⁶ She introduced testimony from four experts—three engineers and a specialist in epidemiology and biostatistics. Making this testimony mesh was essentially a matter of systems analysis, and one of the engineering experts explicitly “employed standard scientific systems analysis techniques . . . ‘[which took into account] all aspects of the automotive system through the full crash sequence.’”⁵⁷

Complex Systems

The increasing significance of a systems approach to engineering reflects increasing complexity and interaction between products and product components. An article by Robert Lucky cites Samuel Arbesman to explain that there are three main reasons for increasing complexity: accretion, interconnection, and edge cases. Accretion, Lucky writes, “is the result of large systems being built on top of smaller and older systems, often via the incorporation of legacy code, producing what we call kludges.”⁵⁸ He further explains that when “these subsystems become interconnected, the resulting entanglement can change what was simply intricate to truly complex.”⁵⁹ “Complexity,” he writes, “is exacerbated by the inevitable existence of edge cases—that myriad of individually negligible exceptions and rarities that yet constitute the long tail of cases that must all be accounted for in system design. The complexity resulting from these factors has passed a tipping point where no single person can fully understand a

53. 975 F. Supp. 2d 485 (M.D. Pa. 2013).

54. Id. at 488–89.

55. Id. at 489.

56. Id. at 500.

57. Id. at 492. See infra note 158 and related discussion.

58. Robert W. Lucky, Should Engineers Start Thinking Like Field Biologists?, IEEE Spectrum, Dec. 20, 2016, available at https://perma.cc/UN6V-4VYJ.

59. Id.

complete system.”⁶⁰ Complexity may also result from hidden causes, as well as from intentionally introduced design parameters.

Complexity also leads to a less understood phenomenon, namely emergent behavior. As discussed by Anderson,⁶¹ when many objects interact within a system, their overall behavior can no longer be understood using a reductionist approach. As the number of engineered devices multiply, such as within the Internet of Things (IoT), no amount of advanced modeling could a priori predict the devices’ ultimate performance. In such a case, there is no substitute for monitoring the actual system in its operational state and designing conservative guardrails. Complexity thus may lead to design flaws that escape detection in the engineering design process.⁶²

Ever since the advent of computing, simulation tools have been used in the design process. While historically, engineering products were designed “by hand,” computers are now ubiquitous as aids to the engineer, and computing tools are used throughout the design and manufacturing processes. The more complicated a system, the more insight may be gained into its behavior by using digital computer simulations. While some systems (such as chaotic systems) may defy causal and deterministic predictions under realistic conditions (slight variations in components), simulations still provide a relatively inexpensive way to model and test engineered components and systems. As more sophisticated tools become available, and with advances in the standard of practice, designers and engineers can model more realistic scenarios, allowing them to push the envelope of performance. Most recently, the availability of more powerful hardware, such as graphical processing units (GPUs), have made machine learning tools more widely available and as such have advanced software development and systems modeling. Below, we discuss two cases where engineering systems failed, and explain the lessons learned.

When Systems Fail

Many cases in which engineers testify arise because an engineered product, structure, or system has failed. Examples of such cases are discussed below in the section titled “Analysis of the Case Law Based on an Understanding of Engineering and How It Differs from Science.” Here we discuss, from an engineering

60. Id.

61. P.W. Anderson, More Is Different: Broken Symmetry and the Nature of the Hierarchical Structure of Science, 177 Science 4047, 393–96 (Aug. 4, 1972), https://doi.org/10.1126/science.177.4047.393.

62. Ch. 3–3: The Role of Failure in Engineering Design: Case Studies, in Intro to ME: Design and Analysis (2003), available at https://perma.cc/5CE7-3EV4 (see https://web1.eng.famu.fsu.edu/~chandra/courses/eml3004c/book/printversion/thebook/0130296333.pdf).

perspective, an aerospace systems failure associated with batteries in a Boeing airplane.⁶³

Among the various engineering challenges Boeing has faced in recent years is the grounding of the Dreamliner fleet because of lithium ion battery fires.⁶⁴

At 10:21 a.m. on [January] 7, 2013, about a minute after all 183 passengers and 11 crew members from Japan Airlines Flight 008 disembarked at Boston’s Logan International Airport, a member of the cleaning crew spotted smoke in the aft cabin of the Boeing 787-8. A mechanic then opened the aft electronic equipment bay of the plane, parked at the airport gate, and saw billowing smoke and flames coming from the batteries for the 787’s auxiliary power unit (APU).⁶⁵ . . . The culprit was a lithium-ion battery manufactured by GS Yuasa. [The battery] was found to be under a condition known as a thermal runaway, in which the heat from a failing cell causes itself and surrounding cells to fail, thereby generating more heat.

[In December 2014], the U.S. National Transportation Safety Board [NTSB] released its report on the Japan Airlines fire. The agency faulted Boeing, GS Yuasa and FAA for shortcomings in their respective roles in the 787 grounding. [This] episode highlights some of the emerging concerns around cutting-edge clean technologies, particularly those that store large amounts of energy. Saving electrons for later is important for applications like electric vehicles, and denser batteries mean vehicles can go farther with less weight, a major consideration for fuel-conscious airlines. Storing electricity is also crucial for smoothing over peaks and valleys in output from wind and solar [generation facilities]. Batteries, especially the lithium-ion variety, are largely filling this role. But doing so at larger scales and with new systems requires additional scrutiny and training for the unique challenges posed and the growing pains that may arise.

63. For information about Boeing, a global aerospace company that “develops, manufactures and services commercial airplanes, defense products and space systems for customers in more than 150 countries,” see Boeing in Brief, Boeing company website, at https://perma.cc/CN8R-CGMS.

64. The description of the Dreamliner batteries incident was reproduced from Umair Irfan, Climatewire (published in Scientific American, December 18, 2014, available at https://perma.cc/4CSK-U2GH).

65. The mechanic used a fire extinguisher to try to put out the flames, “but the blaze didn’t go out. Firefighters arrived at the scene at 10:37 a.m., and using a thermal imaging camera, they looked through the smoke in the equipment bay and saw a softball-sized glow. Responders attempted to extinguish the heat source with Halotron, a fire suppressant. The flames went out, but another firefighter saw that the batteries were still giving off heat and appeared to rekindle. The batteries hissed, leaked fluid and popped. . . . By 12:19 p.m., firefighters had declared the event ‘controlled,’ having removed the APU battery after extinguishing it with further doses of Halotron.” Id.

66. Boeing, supra note 64.

67. AI Engineering A Strategic Research Framework to Benefit Society, Engineering Research Visioning Alliance, 2024, accessible at https://perma.cc/9377-WMDW.

68. See James E. Baker & Laurie N. Hobart, Reference Guide on Artificial Intelligence, p. 1525, in this manual.

principles to a risk equation.” An excellent example of unwitting human bias is described in a study published in Science⁶⁹ in 2019. In that study, it was found that an algorithm widely used in U.S. hospitals to allocate health care to patients has been systematically discriminating against black people—the algorithm was less likely to refer black people than white people who were equally sick to programs that aim to improve care for patients with complex medical needs. The study states that hospitals and insurers use the algorithm and others like it to help manage care for about 200 million people in the United States each year. Engineers may unwittingly succumb to other unknown biases because ML algorithms are based on training data generated and selected by humans. Bias in the training data will lead to biased algorithms. For more detailed discussion of the use of AI in engineering and elsewhere, please see the Reference Guide on Artificial Intelligence,⁷⁰ in this manual.

Analysis of the Case Law Based on an Understanding of Engineering and How It Differs from Science

To gain insight into the kinds of cases in which engineering experts testify and the issues they address, we searched the Westlaw “All Federal” database for all cases decided over the 18 months from January 1, 2021, through June 30, 2022, in which “Daubert,” “Kumho,” or “702” appeared, along with “engineer” or “engineering” in the same sentence as “expert.” Of the 496 cases we found, only 201 (i.e., about 134 cases a year) required a court to address the question of what constitutes valid and reliable engineering testimony.⁷¹ An explanation of how we winnowed down to 201 cases is provided in the footnote below.⁷²

69. Ziad Obermeyer et al., Dissecting Racial Bias in an Algorithm Used to Manage the Health of Populations, Science 366, 447–53 (2019), https://doi.org/10.1126/science.aax2342.

70. See James E. Baker & Laurie N. Hobart, Reference Guide on Artificial Intelligence, section titled “Bias,” in this manual.

71. Of course, the sample did not include the many cases that settle or otherwise resolve without leaving any trace in the Westlaw system.

72. We first eliminated from the Westlaw sample all cases that were not on point. One hundred thirty-eight of 496 cases did not really relate to engineering at all, or were reversed. Of the remaining 358 cases, 76 involved the kind of simple accident investigations with which courts have a long history, and for which judges do not likely require any reference materials beyond what the parties provide. Taking out the 76 simple accident cases from the remaining 358 Westlaw search results left 282 cases for further consideration. The number was reduced yet further by taking out insurance coverage cases. The engineering issues in such cases are confined almost exclusively to causation and qualifications, and the use of experts goes back a long time. In Peele v. Merchants’ Ins. Co., 19 F. Cas. 98, 102 (D. Mass. 1822), the court noted that with regard to the sufficiency of repairs to an insured ship, it would if necessary use “experts to report upon the whole evidence, what in their judgment is the true posture of the case in this respect.” After we removed the

accident and insurance coverage cases, 226 cases remained. Of these, another 25 involved matters like employment related claims, criminal law issues, or civil rights that similarly don’t present difficult engineering evidence admissibility questions. That left 201 cases in which the admissibility challenge was potentially more complex, or about 134 per year. One hundred twenty-six of the 201 cases involved products liability. Twenty-four involved contract or construction litigation claims, 22 involved intellectual property, and 19 involved environmental or land use questions.

73. CV-119-052, 2023 WL 2746033 at *13 (S.D. Ga. Mar. 31, 2023).

74. David G. Owen, Products Liability in a Nutshell 12–13 (10th ed. 2022).

75. Id. at 13.

76. In 1992, the American Law Institute commissioned two professors “to prepare a new Restatement (3d) of Torts on the specific topic of products liability law.” Owen, Products Liability in a Nutshell, supra note 74, at 19. The new Restatement was approved in 1997 and published in 1998, but most jurisdictions have not formally adopted it, and most courts at least nominally still adhere to the traditional products liability legal framework. Id. at 19–20.

77. Id. at 15–20.

tortious misrepresentation, a relatively rarely used theory, “the plaintiff in nearly every products liability case must prove that a product sold by the defendant contained an unnecessary hazard [or defect] that caused [physical] harm,”⁷⁸ and most engineering expert testimony in products liability cases relates to whether there was or wasn’t a defect, and if there was, whether it caused the plaintiff’s injury.

Modern products liability law dates from the late 1950s and early 1960s. Over the past 60 or so years, “courts and commentators [have come] to understand the distinctions between three very different forms of product defect.”⁷⁹ Manufacturing defects are “unintended physical irregularities that can occur during the production process.”⁸⁰ Design defects are “hazards in a product’s engineering or scientific conception that reasonably should be avoided by a different design or formula.”⁸¹ Warning defects occur when there is an “absence of important information about product hazards or how to avoid them.”⁸² Below, we discuss how engineering expert testimony is evaluated with regard to each kind of defect and with regard to the issue of causation. We also discuss engineering testimony on a manufacturer’s duty of care, an issue that arises when a plaintiff claims under a negligence theory.

Design Defects

Depending on the jurisdiction and the facts in a given case, the legal test for determination of a design defect can turn on either (1) what an ordinary consumer might reasonably expect, or (2) a balancing of a product’s risk versus its utility or benefit.⁸³ Plaintiffs generally may introduce expert testimony to establish a defect under either theory, but an expert will more often be necessary to prove a defective design under the risk–benefit test. See, for example, Bradley v. Ameristep, Inc.,⁸⁴ in which the court noted that under Tennessee’s products liability law, a plaintiff invoking the consumer-expectation test can prevail without any expert testimony at all. The consumer-expectation test generally does not apply to complex products, however. As explained in Brown v. Raymond Corp.,⁸⁵ consumers have no reasonable expectations about products so complex they can’t really be understood.

78. Id. at 20, and Restatement (Second) of Torts § 402A.

79. Owen, supra note 74, at 199.

80. Id.

81. Id.

82. Id.

83. See Clayton J. Masterman & W. Kip Viscusi, The Specific Consumer Expectations Test for Product Defects, 95 Ind. L. J. 183, 184–85 (2020), and Ramirez v. ITW Food Equip, Grp., LLC, 686 Fed. App’x 435, 437 (9th Cir. 2017) (consumer-expectations and risk-benefit tests provide alternative means for a plaintiff to prove a design defect). See also Owen, supra note 74, at 186–87.

84. 800 F.3d 205, 211 (6th Cir. 2015).

85. 432 F.3d 640, 643 (6th Cir. 2005).

86. Also, as Professor Owen points out, courts increasingly are turning to the risk-benefit test. David G. Owen, Products Liability in a Nutshell 175 (10th ed. 2022).

87. Brown, 432 F.3d at 641–42.

88. Id. at 647.

89. Id. at 650. See also Lara v. Delta Int’l Mach. Corp., 174 F. Supp. 3d 719, 736 (E.D.N.Y. 2016) (“In a products liability action, ‘the “touchstone” of an expert’s report should be a comparison of the utility and cost of the product’s design and alternative designs.’”); Peck v. Bridgeport Machs., Inc., 237 F.3d 614, 618 (6th Cir. 2001) (reasonable alternative design required); Willet v. Johnson & Johnson, 465 F. Supp. 3d 895, 904 (S.D. Iowa 2020) (“plaintiffs . . . [must] demonstrate the existence of a reasonable alternative design”).

90. Brown, at 643 (plaintiff’s counsel conceded the case depended on applying the consumer-expectation test).

91. 418 F. Supp. 3d 1212 (M.D. Ga. 2019).

point. The plaintiff in the case suffered burns when the lid of a pressure cooker unexpectedly popped off. The court admitted testimony from her expert, who had “a Ph.D. in engineering and . . . extensive experience in design and development of products, fasteners, machines, tools, and latching and locking devices.”⁹² He had tested a pressure cooker like the one that injured the plaintiff, and explained his testing in detail.⁹³ “His familiarity with pressure cookers manufactured by defendant gained by testifying as an expert in similar litigation further demonstrate[d] his experience and knowledge of the product.”⁹⁴ The expert had not proposed any alternative design, but the court held that “it is not required that either plaintiff present evidence of reasonable alternative designs or that the evidence be in the form of expert testimony.”⁹⁵ Though it did not involve a proposed alternative design, Williams falls squarely within the kind of analysis suggested in this reference guide because the issue was the expert’s analysis of an existing design.⁹⁶

Is testing required to prove existence of a defect or the validity of an alternative design?

The question of testing arises in many products liability cases, both to establish a defect and to validate a reasonable alternative design. The focus on testing usually derives from Daubert, not Kumho Tire, and while courts sometimes do suggest testing is part of engineering practice as well as science, they also may confuse or miss the distinction between scientific testing of hypotheses and engineering analysis and testing of designs. For example, in Colon ex rel. Molina v. BIC USA, Inc.,⁹⁷ the court held that “[w]hile testing is not an ‘absolute prerequisite’ . . . it is usually critical to show that an expert ‘adhere[d] to the same standards of intellectual rigor that are demanded in their professional work.’”⁹⁸ The court then discussed what scientists typically do, stating that “[a]dherence to engineering standards of intellectual rigor almost always requires testing of a hypothesis.”⁹⁹

92. Id. at 1222.

93. Id.

94. Id.

95. Id. at 1229.

96. Cf. Bullock v. Volkswagen Grp. of Am., Inc., 107 F. Supp. 3d 1305, 1315 (M.D. Ga. 2015) (court more or less reasoned in reverse to hold that testimony about a reasonable alternative design was sufficient to establish a design defect, even though the plaintiff’s expert was reluctant to describe the problem he saw as a design defect).

97. 199 F. Supp. 2d 53 (S.D.N.Y. 2001).

98. Id. at 76 (quoting Cummins v. Lyle Indus., 93 F.3d 362, 369 (7th Cir. 1996)).

99. Id. (emphases added).

While the decision may well have been sound, it appears to conflate science and engineering.¹⁰⁰

Nease v. Ford Motor Co.¹⁰¹ provides an example of testing to establish if there really was anything at all wrong with the product in question. The plaintiffs claimed that a cable housing design was defective because it allowed dirt and other contaminants to interfere with the operation of a pickup truck accelerator cable. They claimed a crash had resulted when a stuck cable prevented the truck in question from decelerating after the driver let up on the gas pedal. The trial court admitted the testimony of the plaintiffs’ expert, an electrical engineer, but the Fourth Circuit reversed, holding that the expert’s analysis of the design was inadequate. “Testing was of critical importance in this case . . . [but the expert had] conducted no testing whatsoever to arrive at his opinion. . . . [He had not determined] whether it is actually possible for enough debris to accumulate in the casing cap during normal operation to resist the . . . force exerted by the return springs to pull the throttle closed.”¹⁰²

By way of contrast, in Lapsley v. Xtek, Inc.,¹⁰³ an expert’s opinion on a design defect was admitted without testing. The plaintiff was severely injured when a high-intensity jet of grease spurted from a bearing on a huge steel rolling mill machine. His expert explained how the design of the bearing lubrication system differed from a design used on similar rolling mills manufactured both before and after the accident. The expert also explained how this difference caused grease

100. Cf. Amica Mut. Ins. Co. v. Electrolux Home Prods., Inc., 440 F. Supp. 3d 211, 217−18 (W.D.N.Y. 2020) (referring to quintessentially engineering tests as scientific tests while applying Daubert to engineering testimony); Morasch Meats, Inc. v. Ludwick, Case No. DK 18-02714, 2019 WL 5616261, at *2 (W.D. Mich. Sept. 11, 2019) (engineering expert’s testing “is the staple of scientific method, even if his assistance in this matter will not be, strictly speaking, scientific”).

101. 848 F.3d 219 (4th Cir. 2017).

102. Id. at 231–32. See also Elkharroubi v. Six Flags Am. LP, Civ. Action No. TJS-17–2169, 2020 WL 1043304, at *6−7 (D. Md. Mar. 4, 2020) (plaintiff claimed a malfunctioning trap door on a water slide caused his injury, but expert could not “point to a single observation of the trap door mechanism that would render his opinions plausible”); Kesse v. Ford Motor Co., Case No. 14-cv-6265, 2020 WL 832363 (N.D. Ill. Feb. 20, 2020) (theory that electromagnetic interference could cause sudden acceleration was untested); Graves v. Mazda Motor Corp., 675 F. Supp. 2d 1082, 1102 (W.D. Okla. 2009) (In case involving design of automobile gear shift mechanism, expert’s conclusion that shifter was unsafe because it was different from other designs was excluded because it was “not grounded in any objective data or specifically applicable engineering standards. Although [his] Rule 26 report recites his substantial experience with engineering testing in a variety of contexts . . . he did no testing to quantify . . . any exceptional propensity of the gated shifter on the Mazda6 to cause driver confusion about the actual position of the shift lever.”). Cf. Alfred v. Caterpillar, Inc., 262 F.3d 1083, 1088–89 (10th Cir. 2001) (expert should have been allowed to testify that a rotary dial speed control on a paving machine was not in compliance with SAE standards, but his testimony on whether this non-compliance made the machine defective was properly excluded because he had done nothing more than cite to the standard; court affirmed summary judgment for the defendant).

103. 689 F.3d 802 (7th Cir. 2012).

to spurt out. The trial court admitted his testimony, noting that he had employed “commonly known methodologies and physics calculations,” which “included the use of Bernoulli’s equation regarding energy in moving fluids and reference to ‘widely accepted factors concerning the force necessary to penetrate human skin.’”¹⁰⁴ The Seventh Circuit affirmed, holding that “the math and science here are within the comprehension of judges and lawyers without extraordinary assistance. Xtek disputes the completeness and therefore the relevance of [the expert’s] calculations, but it has not identified, and we have not detected, any grave questions about the reliability of the calculations [he] actually performed.”¹⁰⁵

Sometimes an engineer’s experience alone may suffice for admissibility purposes without a requirement for testing of the product at issue or explaining how the analysis of a design supports an expert’s conclusions. In Bullock v. Volkswagen Group of America, Inc.,¹⁰⁶ the plaintiffs claimed a defective turbocharger on a Volkswagen Passat had caused the vehicle to crash when it accelerated uncontrollably. The court denied the defendant’s Daubert motions and allowed a mechanic to testify that a leaking seal had caused the unintended acceleration. He had “investigated and diagnosed unintended acceleration events in cars, and [had] designed a solution to fix a problem with unintended acceleration in prior model Volkswagen turbocharged engines.”¹⁰⁷ He “formed his opinions by observing the physical evidence, testing an exemplar vehicle, and drawing on his extensive experience in the field.”¹⁰⁸ Another expert, a materials engineer, was allowed to

104. Id. at 807–08.

105. Id. at 809–10. See also Russell v. Howmedica Osteonics Corp., No. C06-4078-MWB, 2008 WL 913320, at *5–7 (N.D. Iowa Apr. 2, 2008) (on the question of whether a certain titanium alloy was properly used in a spinal fixator, plaintiff’s expert was allowed to testify based on articles about the relative merits of various metals, even though there was little evidence about when one metal should be used instead of another, or about her opinion that the alloy in question should not be used in patients weighing more than 180 pounds); Dunton v. Arctic Cat, Inc., 518 F. Supp. 2d 296, 300 (D. Me. 2007) (mechanical engineer and a product designer who worked for defendant were allowed to testify about the adequacy and safety of design for snowmobile steering and suspension components); Freitas v. Michelin Tire Corp., No. CIV. 3:94CV1812 (DJS), 2000 WL 424187, at *2 (D. Conn. Mar. 2, 2000) (expert on tire design defect allowed to testify based on “mathematical calculations and experience conducting burst tests on mismatched tires, as well as upon his review of burst tests performed by other reliable testers, sworn testimony and reports of other experts, and industry and tire company documents”). Cf. Shreve v. Sears, Roebuck & Co., 166 F. Supp. 2d 378, 424 (D. Md. 2001) (expert testimony excluded, but “under the unique circumstances of this case, a lay jury could reasonably conclude without the aid of expert testimony that the snow thrower purchased by [plaintiff] was defective [in design] and unreasonably dangerous”); Valente v. Textron, Inc., 932 F. Supp. 2d 409, 426–27 (E.D.N.Y. 2013) (computer simulation not reliable because expert used the wrong coefficient of friction).

106. 107 F. Supp. 3d 1305 (M.D. Ga. 2015).

107. Id. at 1310.

108. Id. at 1311.

testify based only on his experience with automotive component analysis and aviation turbochargers that there were safer seal designs.¹⁰⁹

The three lines of cases: Nease (testing required to prove defect); Lapsley (explanation of analysis is sufficient); and Bullock (reliance on expert’s experience is sufficient) are not inconsistent. All three relate to the question of what caused an unwanted event, and the different legal analyses reflect the fact that different kinds of experts in differing cases can address the question in different ways. Sometimes the answer is evident to a person with adequate experience, and no testing or analysis is required. In such cases, neither the engineering design process nor the scientific method would be implicated. Sometimes the expert has essentially formulated a hypothesis, and admissibility properly would require Daubertlike testing against observational or experimental data. And sometimes an engineering analysis will suffice if it properly reflects the way engineers review and analyze designs. Thus, before deciding on the admissibility of expert testimony, a court needs to first identify what approach the expert has taken, and then apply the appropriate legal analysis.

For plaintiffs in products liability cases, a requirement that an expert must test an alternative (likely by having to build a prototype) can be difficult if their expert has come up with a new design rather than citing to another existing product. In Lara v. Delta International Machinery Corp.,¹¹⁰ the expert testified that a table saw without an interlocking safety device was inherently dangerous, but he conceded that “other than some theoretical musings made during his deposition as to how such a design could be accomplished . . . he [had] never actually designed such an interlock system for use in a table saw.” Nor had he done any testing, which the court held should be part of the “utility versus cost comparison.”¹¹¹ Thus, the testimony was excluded.¹¹²

109. Id. at 1313. See also Great N. Ins. Co. v. BMW of N. Am. LLC, 84 F. Supp. 3d 630 (S.D. Ohio 2015) (engineer with a “range of experiences in vehicle mechanics and design, including some experiential background in aerodynamics, and extensive experience in assessing car defects as it relates to fire defects, and prevention” was allowed to testify on whether design defect was implicated in causing an automobile fire); Sloan Valve Co. v. Zurn Indus., Inc., No. 10–cv–00204, 2013 WL 4052030, at *5 (N.D. Ill. Aug. 12, 2013) (“[Engineering expert] applies his previous education and vast experience to give his expert opinion on whether [defendant’s product] was adequate from a mechanical engineering point of view, thereby aiding the fact-finder in determining the ultimate question of willfulness.”); Bowersfield v. Suzuki Motor Corp., 151 F. Supp. 2d 625 (E.D. Pa. 2001) (plaintiff riding in the rear cargo area of a Suzuki Samurai vehicle was injured when it was hit by another vehicle. Based on experience designing occupant protection systems, plaintiff’s expert was allowed to testify that the Samurai design should have included a seat with a seatbelt in the cargo area or a barrier to keep passengers from sitting there; but expert was not allowed to testify about sufficiency of warnings.).

110. 174 F. Supp. 3d 719, 736 (E.D.N.Y. 2016).

111. Id.

112. Id. at 738. See also Colon ex rel. Molina v. BIC USA, Inc., 199 F. Supp. 2d 53, 75–78 (S.D.N.Y. 2001) (requiring testing of alternative design); Wright v. Case Corp., 2006 WL 278384,

at *4 (N.D. Ga. Feb. 1, 2006) (expert conceded he had done no testing of alternative design, and court held that “testing is particularly important in alternative design cases . . . where it is necessary to show that an alternative design is actually feasible”); Peck v. Bridgeport Mach., Inc., 237 F.3d 614, 618 (6th Cir. 2002) (testimony inadequate where expert said he would have designed a lathe differently, with a “safer” lever shift mechanism, but “had never fabricated such a design and, in fact, had never seen his proposed design used on a lathe anywhere”). Cf. Crawford v. ITW Food Equip. Grp., LLC, 977 F.3d 1331, 1340 (11th Cir. 2020) (expert testimony admissible where he built and demonstrated the alternative design he proposed for a meat saw).

113. 894 F. Supp. 2d 606 (W.D. Pa. 2012).

114. Id. at 630.

115. Id. at 632.

116. Nos. 8:10–cv–219 (MAD/CFH), 8:12–cv–200 (MAD)/(CFH), 2015 WL 2193796 (N.D.N.Y. May 11, 2015).

117. Id. at *11. See also Schenone v. Zimmer Holdings, Inc., No. 3:12–cv–1046–J–39MCR, 2014 WL 9879924, at *8 (M.D. Fla. July 30, 2014) (Expert allowed to testify about alternative design for artificial hip based on his education and experience, which “underpinned his opinion that [defendant’s acquisition of designs for similar products] provided [defendant] with an available and reasonable alternative design [for the product at issue].”); Thierfelder v. Virco, Inc., 502 F. Supp. 2d 1025, 1030–31 (W.D. Mo. 2007) (admitting opinions of expert with years of experience “in the design of wheels and the dimensions of wheelbases for various kinds of mobile furniture” in case involving injury caused by a cart used for folding chairs; expert did no computer modeling or calculations, but did observe the cart and explained proposed changes to protect the feet of users). Cf. Thomas v. FCA US LLC, 242 F. Supp. 3d 819, 831 (S.D. Iowa 2017) (in a case involving allegedly defective airbag, experts did not have to design an alternative airbag because, in their view, defendant already had viable alternatives); Nicholson v. Biomet, Inc., 537 F. Supp. 3d 990, 1010 (N.D. Iowa 2021) (existing products could serve as reasonable alternative design); Borsack v. Ford Motor Co., No. 04 Civ. 3255(PAC), 2007 WL 2142070, at *9 (S.D.N.Y. July 26, 2007) (expert opined there were three alternative door latch designs without citing any testing, though the court did not address whether they were all already in use).

actually go about creating and evaluating designs. Based on that standard, testing is not an absolute requirement, nor is it necessary in many cases. Prototypes are not always required before executing a design, and while moving from design to manufacture of a product may require testing of certain components, that does not mean establishing reasonable feasibility requires such testing. The result in the Lara case (exclusion of expert testimony on alternative design for a power saw) might well have been different had the expert fully explained how the proposed alternative design could feasibly have been executed.

Manufacturing Defects

In manufacturing defect cases, the problem with the product is often quite evident, and close gatekeeping scrutiny is not required. For example, in Schmude v. Tricam Industries, Inc.,¹¹⁸ the parties agreed a ladder was defective because of an improperly installed rivet, but the defendant nonetheless challenged the plaintiff’s expert based on a lack of testing and peer review. The court held that the “defect, by its very nature, did not lend itself to the kind of empirical testing Daubert envisioned for opinions that are based on cutting edge science. . . . Once [the ladder] failed, there was nothing to test. . . . Other than carefully examining the failed rivet, including the portion embedded in the backing plate, there seemed little else for an expert to do.”¹¹⁹

Other cases have subjected engineering experts to closer scrutiny, especially when testing is reasonably possible and would provide important information. In Myrick v. U.S. Saws Inc.,¹²⁰ the plaintiff was injured when a power saw blade manufactured by the defendant broke apart. The plaintiff’s expert attempted to eliminate all possible causes except for a manufacturing defect, and eventually concluded the failure had occurred because of a defective weld. The court rejected this approach,

118. 550 F. Supp. 2d 846 (E.D. Wis. 2008).

119. Id. at 852. See also Martin v. Apex Tool Grp., LLC, 961 F. Supp. 2d 954, 960−61 (N.D. Iowa 2013) (expert’s opinion that flaws in pry bar caused it to fail held admissible based on detailed examination of the bar—which led him to conclude failure originated at location of flaw—size of flaw, location of flaw, the way bar was used, and absence of evidence of misuse; court cited Kumho Tire for proposition that “an expert might draw a conclusion from a set of observations based on extensive and specialized experience”); McCloud v. Terex Telelect Inc., No. Civ–08–433–D, 2010 WL 3259823, at *3 (W.D. Okla. Aug. 17, 2010) (expert engineering testimony on manufacturing defect in step on rear of a truck admitted because expert “testified in his deposition . . . that he employed a recognized method of engineering analysis in reaching his conclusions”); Engler v. MTD Prods., Inc., No. 13–CV–575 (CFH), 2015 WL 900126 at *2 (N.D.N.Y. Mar. 2, 2015) (plaintiff was thrown off a virtually new riding mower after the brakes failed because of excessive wear; engineer’s testimony—that brakes likely had not been adjusted right when the mower was sold—was admitted even though he did not know what had caused the brakes to come out of adjustment or why they were prematurely worn).

120. No. C11–1837Z, 2013 WL 766192 (W.D. Wash. Feb. 28, 2013).

holding that the expert’s testimony was not reliable. Expert “was aware of multiple destructive and non-destructive tests available that could have been done in this case, but [he] conducted no tests on the weld or blade at issue.”¹²¹

Pugh v. Louisville Ladder, Inc.¹²² provides an excellent example of how courts sometimes consider both empirical testing and analytical explanation in assessing engineering testimony on a manufacturing defect. The plaintiff, injured when he fell from a ladder, attributed the accident to manufacturing defects that caused the ladder to fail. In particular, he claimed “microscopic cracks at the ladder’s rivets . . . [had] propagated into larger cracks causing catastrophic failure/buckling that resulted in [his] fall.”¹²³ The defendant ladder manufacturer maintained that the buckling occurred when the plaintiff slipped and fell on top of the ladder.¹²⁴ In other words, the case came down to whether cracks and buckling caused the fall or whether the fall caused the buckling. The jury returned a plaintiff’s verdict, and the defendant appealed, arguing, among other things, that the trial court had abused its discretion in admitting testimony from the plaintiff’s engineering experts. The Fourth Circuit affirmed after rejecting the defendant’s claim that “principles of physics would disprove [plaintiff’s] experts’ theory.”¹²⁵ The plaintiff’s experts initially had concluded the ladder failed based solely on a visual inspection, but thereafter they “performed several tests to support their initial assessment.”¹²⁶ Based on this testing and their experience, they “determined that their structural failure theory was scientifically supported by the facts of this case and the most likely cause of the accident.”¹²⁷ They also performed “testing and analysis to disprove the opposing theory [of] impact damage.”¹²⁸

In many cases, design defect and manufacturing defect claims are alleged together, and the distinction sometimes gets blurred in the admissibility analysis. In Tillman v. C.R. Bard, Inc.,¹²⁹ for example, the plaintiff claimed that a defective filter had been implanted in one of her veins to keep blood clots from migrating

121. Id. at *5. See also Easterling v. Ford Motor Co., 303 F. Supp. 3d 1211, 1222–23 (N.D. Ala. 2018). Plaintiff claimed a seat belt had come unlatched in the course of an accident as a result of a manufacturing defect. His expert engineer’s testimony was excluded because expert admitted he did not know why the buckle had come undone, nor had he done any testing.

122. 361 F. App’x 448 (4th Cir. 2010).

123. Id. at 450.

124. Id.

125. Id. at 453.

126. Id. at 455.

127. Id.

128. Id. While very well reasoned, the court’s opinion did conflate engineering with science, probably because that is how the parties argued the case. For example, the court noted that the plaintiff’s experts “determined that their structural failure theory was scientifically supported by the facts of this case and the most likely cause of the accident.” 361 Fed. App’x at 455. The court cited Kumho Tire once and Daubert 24 times.

129. 96 F. Supp. 3d 1307 (M.D. Fla. 2015).

to where they could cause serious harm. Her doctor planned to remove the device when her clotting risk subsided, but could not do so because it had tilted and become stuck in the vein. Although it had not failed, she claimed that owing to design and/or manufacturing defects it was prone to failure, and that she’d suffered harm because she would be forced to endure monitoring to avoid a failure-related injury. The plaintiff proffered five experts, ranging from an epidemiologist to several engineers, and while the court excluded some of them, all the engineering testimony on manufacturing defects was admitted, even though none of the experts could examine or test the plaintiff’s filter because it remained implanted in her vein.¹³⁰ Whether the design or a manufacturing defect caused this conundrum was not really addressed.¹³¹ The court ultimately denied a motion for summary judgment on the design and manufacturing defect claims simply “because there are issues of fact as to whether the filter’s risks are unreasonable as a result of its arguably defective design or manufacture.”¹³²

Just as courts sometimes do not require distinctly different testimony on design versus manufacturing defect, they also sometimes do not require an expert to specify the exact nature of the defect that caused an injury. In Rudd v. General Motors Corp.,¹³³ the plaintiff was injured while tuning a pickup truck engine. As he leaned over the motor to turn the distributor housing, a blade came off the fan and struck him.¹³⁴ His expert could not pinpoint the cause of the failure, other than to say he thought “the fan blade had to have some small defect for the fatigue failure to start.”¹³⁵ He said that, based on what he had encountered in his experience, there were three possible explanations, but also said there could be others.¹³⁶ He could rule out catastrophic or abnormal use, however, as well as the development of lesser damage that might have occurred post-manufacture.¹³⁷ “Although . . . he did not find direct evidence of a manufacturing defect, such as a microscopic grind mark or inclusion near the fatigue-fracture origin, his testimony [was] replete with circumstantial evidence that—through a process of eliminating alternative explanations—might support a finding of a manufacturing defect.”¹³⁸

130. Id. at 1319–22.

131. Id. at 1354.

132. Id. See also Siegel v. Dynamic Cooking Sys., Inc., 501 F. App’x 397 (6th Cir. 2012) (plaintiff need not establish whether product had either a design or a manufacturing defect).

133. 127 F. Supp. 2d 1330 (M.D. Ala. 2001).

134. Id. at 1340.

135. Id.

136. Id. at 1341.

137. Id.

138. Id. at 1342.

139. Proceeding on both design defect and warning defect claims can create something of an evidentiary conundrum, especially if the plaintiff suggests a reasonable alternative design. If one witness says there’s a serious design defect for which a safer alternative exists, and another says “yes, yes, but the defendant should have warned against the problems caused by the defective design,” the testimony is not completely consistent. Bullock v. Volkswagen Group of America, Inc., 107 F. Supp. 3d 1305 (M.D. Ga. 2015), is an extreme example of the logical problems potentially created by asserting both a defect claim and a warning claim. The court held that even without a warnings expert the plaintiffs could assert a failure-to-warn claim because the defendant knew a defective turbocharger might cause unexpected acceleration. Id. at 1316. Just what good such a warning might do went completely unexplained. The plaintiff did have two experts on the nature of the defect, and apparently neither of them provided an opinion about how a warning might ameliorate it. On the other hand, when a defect can, in fact, be remedied with a warning, the two kinds of defect claims don’t conflict, and in some circumstances the warning claim can proceed on its own. See, e.g., Lara v. Delta Int’l Mach. Corp., 174 F. Supp. 3d 719 (E.D.N.Y. 2016) (in which the court excluded the plaintiff’s design defect expert, but allowed the warning expert to testify).

140. See In re FEMA Trailer Formaldehyde Prod. Liab. Litig., No. MDL 07–1873, 2009 WL 3247967, at *2 (E.D. La. Oct. 6, 2009) (court held there was no need for expert testimony on “the issue of the adequacy of a particular warning” because such a conclusion “involves a common sense assessment ‘within the realm of the average juror’s knowledge and experience’”); McSwain v. Sunrise Med., Inc., Civ. Action No. 2:08cv136KS-MTP, 2010 WL 200004, at *8 (S.D. Miss. Jan. 14, 2010) (plaintiff proffered expert testimony that based on expert’s experience, he knew “that a warning affixed to a product is more effective than a warning inside a manual”; court excluded the testimony because it was “well within the grasp of jurors who have undoubtedly experienced both types of warnings in their daily lives”). Cf. Bradley v. Ameristep, Inc., 800 F.3d 205, 213 (6th Cir. 2015) (because consumer-expectations test applied in case about treestand ratchet straps, plaintiff “was entitled to have a jury employ its own sense of whether the relevant warnings provided the ordinary consumer with knowledge of how to effectively avoid or minimize the risks associated with the . . . straps”).

141. 502 F. Supp. 2d 1025 (W.D. Mo. 2007).

142. Id. at 1027.

143. Id. at 1028–29.

that the “warning label was inadequate in color and wording [because it did not specify] the hazard, how to avoid the hazard and the consequences in not avoiding the hazard of the casters rolling over the feet during use.”¹⁴⁴ The court allowed him to testify on both defect and warnings, and on the latter held that he “was not talking ‘off the cuff’ but rather contrasting the existing ‘Notice’ on the cart with the ANSI standard and dozens of warnings he has previously drafted for other furniture himself.”¹⁴⁵

Still other courts allow testimony from engineers on the need for a warning, but not on the substance, appearance, or placement of the warning. In Pineda v. Ford Motor Co.,¹⁴⁶ for example, the plaintiff was an automotive repair technician who was injured while installing a replacement lift gate on a sport utility vehicle. His only expert, an engineer, testified that the manufacturer’s service manual should have laid out better step-by-step instructions for doing the installation, and that the “manual should have contained an explicit warning [about the need to follow those instructions].”¹⁴⁷ In reversing the trial court’s exclusion of this testimony, the Third Circuit held that while the engineer might not have been the best expert on warnings, his opinions should have been admitted, including testimony about a later warning issued by the defendant, as an example of what kind of language would be acceptable.¹⁴⁸ The court noted that the expert “did not purport to opine on how the warning should be worded or how it should appear in order to effectively convey its message to an automobile technician. He only testified that . . . a warning was necessary to alert a technician to the potential problem.”¹⁴⁹ Neither the engineering process nor the scientific method has much bearing on determining the admissibility of warning testimony in a case like Pineda.

144. Id. at 1029.

145. Id. at 1032. See also Tassin v. Sears, Roebuck & Co., 946 F. Supp. 1241, 1253 (M.D. La. 1996) (“The Court does not believe that a design engineer with experience in both product design and in preparing manuals and warnings must scientifically test an alternative warning before he can opine that a warning is defective.”); Weese v. Black & Decker, Inc., No. 05-3031-CV-S-RED, 2007 WL 4618526 at *1 (W.D. Mo. Sept. 11, 2007) (Expert allowed to testify on both design and warnings where he had “drafted instruction manuals ‘forty or fifty times, probably.’ He wrote warnings for products he designed, and he wrote warnings that are contained in service manuals.”).

146. 520 F.3d 237 (3d Cir. 2008).

147. Id. at 245.

148. Id. at 245–46.

149. Id. at 245. See also Miller v. BGHA, Inc., Civ. Action No. 19–129, 2021 WL 2681277, at *3 (E.D. Pa. June 30, 2021) (the court cited Pineda and held that an “expert need not ‘opine on how the warning should be worded or how it should appear’ to testify that the instructions might result in failure or that a warning is necessary”); Schaaf v. Caterpillar, Inc., 286 F. Supp. 2d 1070, 1074 (D.N.D. 2003) (expert allowed to testify about lack of warnings but not about “the appropriate size, shape, color, content or location of any such warning signs”); Furry v. Bielomatik, Inc., 32 Fed. App’x 882, 883 (9th Cir. 2002) (safety engineer could give both design and warning opinions at conceptual level, without details).

150. 689 F.3d 802 (7th Cir. 2012).

151. The reverse, however, was not true. Proof of the cause did not establish a defect because without showing there was a reasonable alternative design, the plaintiff would have lost. In fact, his expert explained that a design actually used by the defendant both before and after the accident would have prevented the spurting grease, or at least reduced its velocity (Lapsley, 689 F.3d at 807), which is why the plaintiff prevailed. See also Williams v. Tristar Prods., Inc., 418 F. Supp. 3d 1212 (M.D. Ga. 2019) (pressure cooker lid popped off); Bullock v. Volkswagen Grp. of Am., Inc., 107 F. Supp. 3d 1305 (M.D. Ga. 2015) (defective turbocharger caused uncontrollable acceleration). Cf. Lara v. Delta Int’l Mach. Corp., 174 F. Supp. 3d 719 (E.D.N.Y. 2016) (injury clearly caused by saw; court rejected design defect expert but admitted testimony from failure-to-warn expert).

152. 848 F.3d 219 (4th Cir. 2017).

153. Id. at 231–32.

154. See also Elkharroubi v. Six Flags Am. LP, Civ. Action No. TJS-17–2169, 2020 WL 1043304 (D. Md. Mar. 4, 2020) (finding no evidence of defect in trap door on a water slide entailed finding no causation). Cf. Colon ex rel. Molina v. BIC USA, Inc., 199 F. Supp. 2d 53 (S.D.N.Y. 2001) (although cigarette lighter clearly caused plaintiff’s injuries, without evidence of a reasonable alternative design there was no product defect, and hence no cause of action).

155. Case No. 4:13-cv-00800-SRC, 2020 WL 5594059 (E.D. Mo. Sept. 18, 2020).

156. Id. at *5.

but her medical causation opinions were excluded because she was not a medical doctor.¹⁵⁷

Occasionally, the mix of engineering and medical experts occurs in the context of a non-medical product case. Campbell v. Fawber,¹⁵⁸ which involved injuries allegedly caused when the roof of an SUV crushed in an accident, provides an outstanding example of how engineers can address complex causation questions in the context of products liability litigation. The first of three engineering experts in Campbell had a degree in physics, but taught engineering at the National Crash Analysis Center at George Washington University.¹⁵⁹ He was allowed to testify about both the defective design of the SUV and how the structural failure of the roof caused the plaintiff’s injuries.¹⁶⁰ The second expert had a Ph.D. in applied mechanics and had “taught courses on engineering physics and mechanics, and strength of materials, and limit analysis of structures.”¹⁶¹ He offered an opinion on the strength of the SUV’s roof, and the vehicle’s propensity to roll over.¹⁶² The third engineering expert was a biomedical engineer who had “taught courses in the areas of applied mechanics, materials science, biomechanics, bio-materials, ergonomics, occupational health and safety, strength of materials, and orthopaedic engineering.”¹⁶³ He testified that the manufacturer could have designed the roof so the amount of collapse would have been greatly reduced, and that a poorly designed seat belt may also have contributed to the plaintiff’s injuries.¹⁶⁴

In explaining its decision to admit the testimony of all three engineers, the Campbell court went to some length to distinguish their work from that done by experts in other cases who were excluded. The first engineering expert “uses established systems analysis techniques to form his opinion, evaluating ‘crash statistics, vehicle dynamics, occupant kinematics, injury susceptibility and mechanism, crashworthiness, and occupant protection.’ He examines a number of materials, including those published in peer-reviewed journals. Most importantly, however, [he] relies on [information about other vehicles with a roof structure similar to the SUV, and with alternative designs].”¹⁶⁵ The second expert reviewed extensive information and concluded that “the roof intrusion over [the plaintiff’s] seat would have subjected her to as much as twice [the] dynamic deflection as her body could withstand.”¹⁶⁶ Similarly, the third expert “observed the crash location,

157. Id. at *6.

158. 975 F. Supp. 2d 485 (M.D. Pa. 2013).

159. Id. at 491.

160. Id. at 492.

161. Id. at 493.

162. Id.

163. Id. at 494.

164. Id.

165. Id. at 498.

166. Id. at 499.

and took measurements in order to reconstruct the accident scene [and explained that the SUV had a higher-than-acceptable propensity to roll over].”¹⁶⁷

The defendant argued that no expert had conducted a test of a modified SUV to show that remedying the supposed defects would have prevented the plaintiff’s injuries. The court rejected this rote invocation of testing. To require such tests would make plaintiffs “prove their case twice.”¹⁶⁸ The court also noted that the factors listed in Daubert do not constitute a definitive checklist or test for admissibility.¹⁶⁹ Thus, Campbell is another example of how engineering design evaluation rather than scientific testing of hypotheses can be the basis for admissible causation testimony.

There are cases, however, in which causation testimony from an engineering expert is properly analyzed based on hypothesis testing criteria. Many fire investigation cases fall into this category. Werth v. Hill-Rom, Inc.¹⁷⁰ is an excellent example that also involved products liability. The plaintiff was a newborn infant who required oxygen therapy. He was placed under an oxygen hood in a bassinet with a baby warmer over him.¹⁷¹ When a fire erupted in the bassinet, the hospital and his parents blamed it on “a hot particle [that] had broken from the baby warmer and fallen into the bassinet, igniting the flammable materials (blankets and the like) below,”¹⁷² but the court excluded expert testimony on this theory of causation. Citing a publication from the National Fire Protection Association, it found that “once an investigator has developed a hypothesis about a fire’s cause, [a key step] is testing that hypothesis. Yet, nothing in the Report or the Supplemental Report indicates that the team tested its hypothesis at all. Indeed, as [one investigator] candidly acknowledged in his deposition, it never did.”¹⁷³

Duty of Care

In products liability cases based on the legal theory of negligence rather than strict liability or warranty, a plaintiff has to establish that the defendant breached a duty of care, and engineers sometimes testify about the duty of care as well as product defect or causation. In Crawford v. ITW Food Equipment Group, LLC,¹⁷⁴ the plaintiff lost an arm “when it came into contact with the unguarded blade of one of [the defendant’s] commercial meat saws.”¹⁷⁵ His engineering expert testified about

167. Id.

168. Id.

169. Id.

170. 856 F. Supp. 2d 1051 (D. Minn. 2012).

171. Id. at 1053.

172. Id.

173. Id. at 1060.

174. 977 F.3d 1331 (11th Cir. 2020).

175. Id. at 1336.

alternative designs of which the defendant should have known, and this testimony sufficed to establish negligent design.¹⁷⁶ Obviously, testimony about what the defendant should have known is a matter of engineering knowledge, and Daubert’s admissibility criteria for scientific evidence would not apply.

Non-Products Liability Cases

Contract and Construction Litigation Cases

Engineers are often called as experts when disputes arise about contract compliance, especially in the context of construction contracts.¹⁷⁷ For example, Waste Management of Louisiana, L.L.C. v. Jefferson Parish¹⁷⁸ involved a complex dispute between Jefferson Parish, Louisiana, and Waste Management, the company that operated, managed, and maintained the parish’s sanitary landfill.¹⁷⁹ The core issues in the case related to (1) what work the plaintiff, Waste Management, and its subcontractor and counterclaim plaintiff, T & K Construction, were supposed to do, (2) when it was to be done, and (3) how the parish might have interfered with that work. The parish retained a civil engineer to serve as an expert on these issues, and the two plaintiffs moved to exclude his testimony.

The plaintiffs did not challenge the expert’s methodology or theories, or the data he considered, “except to the extent that they challenge his interpretations of provisions of the [contract with Waste Management].”¹⁸⁰ Though the court recited the Daubert factors, it made no real attempt to apply them, and ultimately held that the expert had “explained his methodology, which involved a careful review of all of the relevant pleadings and documents provided by the parties, as well as an interview with the Landfill engineer and a visit to the Landfill. [His] expert report reveals that he has more than 30 years of engineering experience across the fields of sewerage systems, site engineering, land surveying and structural engineering, among others, albeit not seemingly focusing on landfills. [Though he] may lack expertise on certain questions of landfill construction, . . . his overall experience suggests that Waste Management and T & K have failed to set forth that any such failing disqualifies [his] expert report in its entirety.”¹⁸¹

The court was concerned, however, that the expert had strayed from engineering testimony into contract interpretation. “The Court agrees that [the]

176. Id. at 1343–44.

177. For a good discussion of construction industry Daubert cases, see C.J. Heffernan et al., Defending and Asserting Daubert Challenges in Construction Disputes, 32 The Construction Lawyer (Spring 2012).

178. Civ. Action No. 13–6764, 2015 WL 5798029 (E.D. La. Oct. 5, 2015).

179. Id. at *1.

180. Id. at *12.

181. Id.

expert report is rife with legal conclusions and contractual interpretations. . . . However, the Court finds that although many of the opinions offered in [the] expert report are inadmissible, [the expert] . . . may testify that ‘[a]s a national landfill operator, Waste Management should have been aware of the problems that gas could cause in capping work and should have made provisions for it in the plans they developed.’”¹⁸²

Waste Management is a prime example of how the Daubert factors do not properly apply to some engineering testimony, and other contract cases further illustrate this point. In First Assembly of God Church v. Fondren,¹⁸³ the issues were whether a construction company had complied with drawings and specifications, and whether failure to comply had resulted in the collapse of a church’s roof. The defendant raised about every argument possible in unsuccessfully challenging the plaintiff’s engineering expert.¹⁸⁴ On the facts of the case, however, the broadside attack on the expert verged on the frivolous. The expert was “a registered public engineer in nine states . . . and [had] actively practiced engineering for [about 29 years].”¹⁸⁵ He had “previously conducted investigations into numerous other building design and construction defects as well as building collapses,” co-authored articles, and “testified in numerous cases.”¹⁸⁶ Here, the expert opined that “[b]ased on my knowledge of the behavior of rigid frame metal buildings, I determined that braces should have been installed at the knee joints and that braces would have precluded the observed failure.”¹⁸⁷ The defendant did not raise “any specific objections to the methodology used by [the expert] . . . [and made] no argument as to why additional calculations on the bending strength of the steel frame [were] necessary.”¹⁸⁸

182. Id. at *13.

183. No. 5:02CV47, 2003 WL 25685226 (E.D. Tex. Jan. 8, 2003).

184. Id. at *3. There were eight supposed grounds for the motion to exclude: “(1) he is not qualified to give the opinions; (2) his theories have not and cannot be tested; (3) his testimony is based upon subjective interpretation rather than valid theory; (4) his theories have not been subjected to proper peer review and/or publication; (5) his opinions have a potential high rate of error; (6) he is attempting to testify to matters which are not scientifically correct and with no scientific basis for his opinions; (7) his opinions and conclusions should be excluded because they are incorrect, and the analysis used to reach the opinions and conclusions are not reliable; and (8) he is not a licensed engineer in Texas.”

185. Id. at *5.

186. Id.

187. Id. at *6. Cf. Allegra v. Luxottica Retail N. Am., 341 F.R.D. 373, 438−42 (E.D.N.Y. 2022) (plaintiffs claimed defendant manufacturer of eyeglasses had made deceptive misrepresentations about the significance of the precision with which its lenses were made; court held an engineer could testify for plaintiffs based on his review of literature about the lens-making machines at issue).

188. Id. See also United States ex rel. Poong Lim/Pert Joint Venture v. Dick Pac./Ghemm Joint Venture, 2006 WL 5230015, at *3 (D. Alaska Mar. 2, 2006) (in a dispute between a contractor and subcontractor, the court rejected the plaintiff’s motion to exclude the defense experts, holding that in “cases where the subject of the expert testimony is non-scientific, as the case at bar, education,

training and experience, not necessarily methodology or theory, are more appropriate benchmarks for determining reliability”); Ferguson v. Erie Ins. Prop. & Cas. Co., Civ. Action No. 3:19-0810, 2021 WL 4392040, at *3 (S.D. W. Va. Sept. 24, 2021) (in dispute about what caused walls to crack, expert allowed to testify “as to his own personal observations and experience as an engineer”).

189. Civ. Action No. 2:18-cv-83-KS-MTP, 2020 WL 8254812 (S.D. Miss. Dec. 1, 2020).

190. Id. at *2.

191. Id.

192. Id. at*2–*3.

193. Id. at *3.

194. 514 F. Supp. 2d 439 (S.D.N.Y. 2007).

195. Id. at 441.

beats from [the allegedly infringed recording].”¹⁹⁶ The court granted summary judgment after finding that the expert had failed “to substantiate his opinion. . . . He offers no explanation for what the tools [for sound manipulation] are; provides no significant detail regarding how they alter sound; and provides no demonstration of how a sound from [plaintiff’s recording] would be manipulated to achieve one of the sounds in [the other recording].”¹⁹⁷ The engineering issue in cases like Vargas would be how engineers develop and use tests to determine similarity of recordings.¹⁹⁸

Engineers in trademark cases

Engineers sometimes appear in trademark cases, though the experts more typically are people who conduct surveys to see if one party’s use of a logo or the like has caused confusion with another party’s logo or the like. In Malletier v. Dooney & Bourke, Inc.,¹⁹⁹ one issue was whether the defendant’s use of certain colors caused confusion about multicolor handbag logos.²⁰⁰ The court denied a motion to exclude an expert on colorimetry who was “a principal engineer in color technology, and a professor at Boston University, researching and teaching issues pertinent to vision and color.”²⁰¹ Again, the engineering issue would be the development and use of tests.

Engineers in patent cases

Most engineering expert testimony on intellectual property issues occurs in patent litigation. In the United States, patents are issued by the U.S. Patent and Trademark Office (USPTO) after an extensive application and examination process.²⁰² There are three kinds of patent. A utility patent “may be granted to anyone who invents or discovers a new and useful process, machine, article of manufacture, or composition of matter, or any new and useful improvements of

196. Id. at 445.

197. Id.

198. See also SA Music LLC v. Apple, Inc., 592 F. Supp. 3d 869, 900–02 (N.D. Cal. 2022) (excluding testimony from expert software engineer who specialized in audio and music technology).

199. 525 F. Supp. 2d 558 (S.D.N.Y. 2007).

200. Id. at 570.

201. Id. at 641.

202. See Peter S. Menell et al., Patent Case Management Judicial Guide, Third Edition (Federal Judicial Center 2016), pp. 14-9 to 14-29.

these.”²⁰³ A utility patent may cover a business method.²⁰⁴ A design patent may be granted to “anyone who has invented a new, original ornamental design for an article of manufacture.”²⁰⁵ A plant patent may be granted to “anyone inventing or discovering and asexually reproducing any distinct and new variety of plant.”²⁰⁶ Utility patents far and away constitute the majority of patents granted,²⁰⁷ and they are the primary focus here.

Every patent includes a specification that describes the patented invention and concludes with what are called claims. Claims “are commonly analogized to the ‘metes and bounds’ of a property deed and serve the same purpose: to delineate the scope of the asset which, in the patent context, is an invention.”²⁰⁸ Importantly, a patent does not give the holder the right to make or sell anything. Rather, each “claim represents the legal right to exclude others from making, using, selling, offering to sell, importing, or offering to import the claimed process, machine, manufacture, or composition of matter.”²⁰⁹ “A patent may, and often does, contain many claims, which usually become increasingly specific” as they progress from first claim to last.²¹⁰

A simple hypothetical illustrates how claims work and what rights they confer if the USPTO grants a patent. Suppose Inventor A has a patent with two claims: (1) a drink container comprising a cylinder closed at one end and open at the other, and (2) a drink container as in Claim One wherein the closed end cylinder is made of ceramic material. Claim One, which would be considered an independent claim, would cover, or “read on” almost any drinking glass, except, for example, a non-cylindrical glass that’s wider at the top than at the bottom. Claim Two, which would be called a dependent claim, would only read on a drinking “glass” made of ceramic material.²¹¹

Now suppose Inventor B comes up with a new idea that would make it easier to use one of the cylindrical drink containers for hot beverages, like coffee.

203. See United States Patent and Trademark Office, “Applying for Patents,” available at https://perma.cc/JY3T-Z83H (accessed Mar. 6, 2024).

204. State Street Bank & Trust Co. v. Signature Fin. Grp., Inc., 149 F.3d 1368 (Fed. Cir. 1998).

205. See USPTO, “Applying for Patents,” supra note 203.

206. Id.

207. Menell, Patent Case Management Judicial Guide, at 1–13.

208. Id. at 14-7 to 14-8.

209. Id. at 14-8 (emphasis added) & 35 U.S.C. § 271(a).

210. Menell, Patent Case Management Judicial Guide, at 14-8.

211. Note that in patent law, the term “comprising” simply means having at least the elements in the claim. See CIAS, Inc. v. Alliance Gaming Corp., 504 F.3d 1356, 1360 (Fed. Cir. 2007) (“In the patent claim context the term ‘comprising’ is well understood to mean ‘including but not limited to.’”), cited in In re Skvorecz, 580 F.3d 1262, 1267 (Fed. Cir. 2009). Cf. Guardant Health, Inc. v. Foundation Medicine, Inc., Civ. Action No. 17-1623-LPS-CJB, 2019 WL 5677748 at *15 (D. Del. Nov. 1, 2019) (making clear that an independent claim must be construed to include a dependent claim).

Inventor B obtains a patent with a single claim: a drink container comprising a ceramic cylinder closed at one end and open at the other and a ceramic handle attached to said cylinder. The claim in this patent would read on a coffee mug, at least one cylindrical in shape. Can Inventor B now start making and selling cylindrical coffee mugs with handles? No, because doing so would infringe on the claims in Inventor A’s patent. Inventor A, however, could not make or sell the mugs either, because doing so would infringe on Inventor B’s patent. Would it then be impossible to make and sell coffee mugs in our hypothetical world? As a practical matter, probably not, because the two patent holders likely would enter into a cross-licensing agreement that would let them both make mugs.²¹²

Patent litigation almost always arises when a patent holder believes a competitor is making or selling something that infringes on the patent, or when the competitor seeks a declaratory judgment that its “accused” process, product, or method does not infringe. Infringement means at least one claim is infringed;²¹³ that is, at least one claim reads on the allegedly infringing process, product, or method. Once litigation has commenced, there are three principal issues on which experts may testify: (1) claim construction, (2) infringement or non-infringement, and (3) validity or invalidity.

Justice Breyer’s opinion in Markman recognized that because a trial court has to consider at least some evidence in addition to the language of the claims, interpretation of a claim is not the kind of purely legal question typically addressed by a trial judge rather than the jury. Construing “a term of art following receipt of

212. For a real-life example of cross-licensing, see Intel Corp. v ULSI System Technology, Inc., 995 F.2d 1566 (Fed. Cir. 1993).

213. Grober v. Mako Prods., Inc., 686 F.3d 1335, 1344 (Fed. Cir. 2012).

214. Menell, Patent Case Management Judicial Guide, supra note 202, at 5-3.

215. J. Jonas Anderson & Peter S. Menell, Informal Deference: A Historical, Empirical, and Normative Analysis of Patent Claim Construction, 108 Northwestern Univ. L. Rev. 1, 3–4 (2014).

216. 517 U.S. 370, 384–88 (1996).

evidence” was termed a “mongrel practice.”²¹⁷ Nonetheless, taking into account precedents in prior cases and “the relative interpretive skills of judges and juries and the statutory policies that ought to be furthered by the allocation [between judge and jury],”²¹⁸ the Court held “there is sufficient reason to treat construction of terms of art like many other responsibilities that we cede to a judge in the normal course of trial, notwithstanding its evidentiary underpinnings.”²¹⁹

Only a few months after Markman, the Federal Circuit, in Vitronics Corp. v. Conceptronic, Inc.,²²⁰ held that a trial judge had erred in relying on expert testimony in interpreting a claim. The Federal Circuit distinguished between intrinsic evidence (the patent file and history) and extrinsic evidence (e.g., expert testimony, inventor testimony, dictionaries, and technical treatises and articles).²²¹ “[E]xtrinsic evidence in general, and expert testimony in particular, may be used only to help the court come to the proper understanding of the claims; it may not be used to vary or contradict the claim language.”²²² Given this rule, expert testimony during the claim construction stage of a litigated case is less frequent than during the trial, when experts often testify about infringement and validity.²²³

Moreover, when experts do testify on claim construction, their opinions typically derive from their experience, and not from any engineering analysis or scientific hypothesis testing. Thus, challenges to the admissibility of their testimony focus mostly on their education, training, and experience. Pentair Water Pool and Spa, Inc. v. Hayward Industries, Inc.²²⁴ is a good example of how courts evaluate such testimony. The case involved heaters for swimming pools and spas. During the Markman hearing, the defendant tried to introduce testimony from a materials engineering professor. The court held the professor’s testimony largely inadmissible because the expert was not someone with adequate “skill in the

217. Id. at 378.

218. Id. at 384.

219. Id. at 390.

220. 90 F.3d 1576 (Fed. Cir. 1996).

221. Id. at 1584.

222. Id.

223. The allocation of claim construction to the judge quickly led to the practice of holding pretrial “Markman hearings.” Such hearings promote settlement and provide a basis for summary judgment. J. Jonas Anderson & Peter S. Menell, Informal Deference: A Historical, Empirical, and Normative Analysis of Patent Claim Construction, 108 Northwestern Univ. L. Rev. 25 (2014). These hearings also facilitate the development of expert reports on the issues of infringement or validity, because they clarify how the claims in a patent have to be interpreted. Id. Because Markman hearings occur well before trial, and sometimes even before discovery, they often involve no expert testimony at all. J. Michael Jakes, Using an Expert at a Markman Hearing: Practical and Tactical Considerations, IP Litigator, Aug. 2002 at 2 (“Once the primary means for presenting claim-construction issues to juries, expert testimony may or may not be allowed by the courts today.”), available at https://perma.cc/GLH3-UXVG.

224. Case No. CV 11-10280-GW(FMOx), 2012 WL 12887686 (C.D. Cal. Dec. 12, 2012).

[relevant] art.”²²⁵ The court found that while he was “clearly a distinguished Professor of Engineering, [he was not] a person of ordinary skill in the art in the field to which the . . . patent pertains. He admitted as much at his deposition where he said ‘I’m not of ordinary skill, but I know who such people are. I teach them and have taught them for years.’”²²⁶

When the actual merits of expert testimony on infringement are addressed, the scrutiny is often less strict than in products liability cases. In Fontem Ventures, B.V. v. NJOY, Inc.,²²⁹ a case involving several patents for e-cigarette devices, the plaintiff’s expert on infringement was a mechanical engineer with “40 years’ experience in the design, development, and manufacture of electromechanical devices, both as an employee and as a consultant.”²³⁰ For each allegedly infringing device he “provide[d] photographs of the accused products annotated with numerals in the ‘Infringement Analysis’ and brackets or arrows pointing to portions of the accused products, referencing numerals from [various exhibits].”²³¹ His conclusion was essentially “look at the patent and look at what the defendants have done.” Neither the engineering design process nor the scientific method was involved.

225. Id. at *11.

226. Id. See also Daiichi Sankyo Co. v. Apotex, Inc., 501 F.3d 1254, 1256 (Fed. Cir. 2007) (listing factors that a court can consider for determining if an expert has the requisite level of ordinary skill in the art at issue).

227. 575 F.3d 1312 (Fed. Cir. 2009).

228. Id. at 1321.

229. CV 14-1645-GW(MRWx), 2015 WL 12743861 (C.D. Cal. Oct. 22, 2015).

230. Id. at *3.

231. Id.

232. Fontem Ventures, 2015 WL 12743861, at *7. See also Gatearm Techs., Inc. v. Access Masters, LLC, Case No. 14-62697-CIV-SCOLA/OTAZO-REYES, 2020 WL 6808670, at *14 (S.D. Fla. Apr. 30, 2020) (defendant’s expert testimony on non-infringement admitted based on his comparison of defendant’s original infringing product with a new design, which he concluded was very different).

233. Fontem Ventures, 2015 WL 12743861, at *8. See also DataQuill Ltd. v. Handspring, Inc., No. 01 C 4635, 2003 WL 737785, at *2–4 (N.D. Ill. Feb. 28, 2003) (in a case involving data entry devices, the court held that “operating and examining the accused devices (assuming it is done against the backdrop of a proper claim construction) is a reliable method of conducting an infringement analysis,” but nonetheless excluded expert’s testimony because he had not properly construed the claims). Cf. Fuma Int’l LLC v. R.J. Reynolds Vapor Co., 1:19-CV-260, 1:19-CV-660, 2021 WL 4820738, at *2 (M.D.N.C. Oct. 15, 2021) (another case involving e-cigarette devices). In Fuma, the court held the plaintiff’s expert testimony admissible to the extent it was “based on technical analysis on several different points. He discusses technical design similarities between Fuma’s patented e-cigarette design and RJR’s Solo and Ciro products, such as a transverse heating element and central airflow passageway. . . . He also uses a technical analysis to explain his opinion that RJR’s argument that Fuma’s e-cigarette design was different than its own is not valid, and that RJR’s portrayal of Fuma’s e-cigarette design could not electrically function.” But see Elder v. Tanner, 205 F.R.D. 190 (E.D. Tex. 2001), in which the court held it “is not sufficient simply to list the resources [experts] utilized and then state an ultimate opinion without some discussion of their thought process.” Id. at 194.

234. Id.

235. Schumer v. Lab. Computer Sys., Inc., 308 F.3d 1304, 1315 (Fed. Cir. 2002) (“Evidence of invalidity must be clear as well as convincing.”).

236. White Paper Report: United States Patent Invalidity Study 2012, available at https://www.morganlewis.com/~/media/files/publication/presentation/speech/smyth_uspatentinvalidity_sept12 (accessed July 14, 2024).

237. Section 102, 35 U.S.C., provides that a person is not entitled to a patent if “(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention; or (2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.”

238. Section 103, 35 U.S.C., provides that a “patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains.”

239. See Therasense, Inc. v. Becton, Dickinson, Case Nos. C04-02123 MJJ, C04-03327 MJJ, C04-03732 MJJ, 2007 WL 2028197 at *3 (N.D. Cal. July 10, 2007) (“A claim is anticipated under 35 U.S.C. § 102 only if each and every limitation of the claim is disclosed in a single prior art reference.”).

240. See Izzo Golf, Inc. v. King Par Golf Inc., 561 F. Supp. 2d 334, 338 (W.D.N.Y 2008) (“A claim is obvious if the patented invention would have been obvious to a person having ordinary skill in the art to which the subject matter of the patent pertains, at the time the invention was disclosed.”) (emphasis in original).

241. Cohesive Techs., Inc. v. Waters Corp., 543 F.3d 1351, 1364 (Fed. Cir. 2008) (“While it is commonly understood that prior art references that anticipate a claim will usually render that claim obvious, it is not necessarily true that a verdict of nonobviousness forecloses anticipation. The tests for anticipation and obviousness are different. . . . Obviousness can be proven by combining existing prior art references, while anticipation requires all elements of a claim to be disclosed within a single reference . . . ‘[I]t does not follow that every technically anticipated invention would also have been obvious.’”) citing In re Fracalossi, 681 F.2d 792, 796 (CCPA 1982) (Miller, J., concurring).

Mobile Medical International Corp. v. Advanced Mobile Hospital Systems, Inc. illustrates how courts analyze the admissibility of engineering testimony on obviousness.²⁴² The plaintiff company in the case brought a declaratory judgment action to invalidate the defendant’s patent for a mobile medical unit. The plaintiff company’s expert was a consultant in the field of specialty vehicle engineering, including medical vehicles. The defendant challenged the expert’s experience and also argued that his opinion on obviousness was based on “hindsight bias.” The court rejected a Daubert motion to exclude the expert, holding on the hindsight issue that his testimony about not using post-patent information sufficed.²⁴³ There was nothing about the application of the Daubert factors for scientific testimony or about the engineering design process.

Meridian Manufacturing, Inc. v. C&B Manufacturing, Inc.²⁴⁴ provides an example of a more detailed analysis of expert testimony on obviousness, but while the court initially discussed Daubert at some length,²⁴⁵ the ultimate decision on two challenged engineering experts was based mostly on the engineering analyses they had done. The invention at issue was an agricultural trailer specially designed for the transport of large bags or boxes of seeds to planters in the field.²⁴⁶ The defendant (alleged infringer) claimed the patent was “clearly an obvious combination of well-known elements from other patents.”²⁴⁷ The court undertook a point-by-point discussion of each expert, and focused largely on their explanations for how they reached their conclusions. For example, the court found that at the beginning of the report prepared by the defendant’s expert, he “discussed generally the geometric considerations and principles of guiding objects onto surfaces that go towards the ‘common sense’ category of motive to combine. While not determinative, and certainly subject to cross-examination and rebuttal, this [was] sufficient to permit [his] obviousness opinion to be admitted into evidence.”²⁴⁸

Engineers in trade secret cases

Engineers also sometimes appear as expert witnesses in trade secret cases. Although the primary fact questions in such cases are usually about who stole or disclosed secret information, defendants sometimes will claim the information wasn’t really a secret because anyone could easily have figured it out. In Raytheon Co. v. Indigo Systems Corp.,²⁴⁹ the defendant’s expert opined that some supposed trade secrets

242. No. 2:07–cv–231, 2015 WL 778553 (D. Vt. Feb. 24, 2015).

243. Id. at *4.

244. 340 F. Supp. 3d 808 (N.D. Iowa 2018).

245. Id. at 829–30.

246. Id. at 823.

247. Id. at 828.

248. Id. at 834.

249. 598 F. Supp. 2d 817 (E.D. Tex. 2009).

could have been reverse engineered.²⁵⁰ He also suggested that other alleged trade secrets did not “meaningfully differ from information that is widely known in the industry.”²⁵¹ The court admitted his testimony with no detailed analysis.²⁵²

Environmental/Land Use Cases

The use of experts in cases about environmental or land use claims goes back almost 250 years, to Folkes v. Chadd,²⁵³ the English case generally considered the first example of allowing experts to provide opinion testimony. Today, engineers often testify in such cases. For example, in Adler v. Elk Glenn, LLC, the court admitted testimony from a geotechnical engineer about the cause of land subsidence under the plaintiff’s property.²⁵⁴ This expert had

consulted mine maps and websites, United States Geological Survey maps, topographic maps, excavated material, and rock strata visible in nearby areas. He also relied on his extensive experience with mine spoil to figure out the likely distribution of materials within the fill beneath the house. . . . He developed this opinion based on his extensive personal experience with mine spoil and his review of scholarly literature finding that mine spoil settlement worsens over time [and he visited the property].²⁵⁵

The fact that he had not performed readily available tests did not make his testimony unreliable because he had “followed the prevailing methods of his profession.”²⁵⁶ Adler is another example of how engineering analysis rather than scientific hypothesis testing can suffice to establish causation.

250. Id. at 821.

251. Id. at 821–22.

252. Cf. Coda Development s.r.o. v. Goodyear Tire & Rubber Co., 565 F. Supp. 3d 979 (N.D. Ohio 2021), in which the court denied a motion to exclude the plaintiff’s expert on misappropriation of trade secret information about self-inflating tires because the defendant’s motion had focused on the expert’s knowledge and experience, and had “not presented a rigorous application of the teachings of Daubert.” Id. at 998. It is unclear how the scientific method might have been relevant.

253. 99 Eng. Rep. 589 (1782).

254. 986 F. Supp. 2d 851 (E.D. Ky. 2013).

255. Id. at 855.

256. Id. at 856. See also Lake Charles Harbor & Terminal Dist. v. Reynolds Metal Co., Case No. 2:17-CV-01114, 2022 WL 838394 at *3 (W.D. La. Mar. 21, 2022) (expert on cleaning up landfills allowed to testify “based on his experience, expertise, and education”).

257. Civ. Action No. 1:13-CV-1375-LMM, 2016 WL 3870080 (N.D. Ga. Feb. 17, 2016).

258. Id. at *1.

259. Id. at *2.

260. Id.

261. Id. at *7.

262. Id. at *11. Cf. Kirksey v. Schindler Elevator Corp., Civ. Action 15-0115-WS-N, 2016 WL 5213928 (S.D. Ala. Sept. 21, 2016) (use of failure modes and effects analysis to determine inadequacy of design for escalator handrail).

263. No. 17C7567, 2020 WL 5570041 (N.D. Ill. Sept. 17, 2020).

264. 769 F. Supp. 2d 269 (S.D.N.Y. 2011).

something called Hazen Units. The defendant questioned whether the plaintiff’s witnesses were “sufficiently expert in ‘chemistry or phenol discoloration.’”²⁶⁵

Occasionally, the validity of a test rather than whether it was properly conducted may be the central issue. A case that involved toxicology testing rather than engineering best illustrates this point. In Ruffin v. Shaw Industries, Inc., the plaintiffs claimed off-gassing from carpets was toxic.²⁶⁶ Their expert had developed her own test to determine toxicity, and the defendant challenged its validity. The trial court excluded her testimony, and the Fourth Circuit affirmed based on an application of the Daubert factors. The Environmental Protection Agency had attempted to replicate the test results, but could not do so.²⁶⁷

Conclusion

The two core messages of this reference guide are (1) the way in which engineering differs from science, and (2) the centrality of the design process and the evaluation of designs to the way engineers think and go about their work. Our review of cases in which engineering expert testimony is at issue shows that courts by and large admit testimony based on engineering design and analysis without requiring the kind of hypothesis testing and peer-reviewed publication that is central to the scientific method. When the testimony is more like science than engineering, however, at least some of the Daubert criteria for the evaluation of science are often applied; and sometimes admissibility depends primarily on an expert’s knowledge and experience rather than either hypothesis testing or engineering analysis. These distinctions are fully in keeping with Kumho Tire’s admonition that the objective in determining the admissibility of expert testimony is “to make certain that an expert, whether basing testimony upon professional studies or personal experience, employs in the courtroom the same level of intellectual rigor that characterizes the practice of an expert in the relevant field.”²⁶⁸

265. Id. at 277.

266. 149 F.3d 294 (4th Cir. 1998).

267. Id. at 297–98.

268. Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137, 152 (1999).