SUMMARY

While considerable progress has been made to address utility issues before a project goes to letting, a substantial knowledge gap remains relative to the management utility conflicts during construction. NCHRP Project 15-69, “Utility Conflict Impacts during Highway Construction,” addressed this issue by identifying causes and impacts of utility issues during construction, evaluating the use of utility impact analysis tools, developing functional requirements for a decision support tool, documenting procedures and tools for conducting utility inspections, and documenting best practices and implementation recommendations.

Research deliverables include this conduct of research report (NCHRP Web-Only Document 396: Strategies to Address Utility Issues during Highway Construction); NCHRP Research Report 1100: Minimizing Utility Issues During Construction: A Guide; presentation materials describing the research background, approach, findings, and conclusions; and an implementation plan.

CAUSES OF UTILITY ISSUES DURING CONSTRUCTION

Literature Review

A literature review focused on practices and issues related to utility-related impacts on project delivery as well as construction and utility inspection practices. Common factors mentioned included utility relocation delays; differing site conditions (DSCs); errors in plans, specifications and estimate (PS&E); inaccurate utility facility locations, and owner-requested changes.

Other than participation in surveys and interviews, the technical literature was scant on the impact of utility-related issues to contractors. Examples that were noted included (a) lower production rates for installing underground appurtenances that were in conflict with existing or unknow utility facilities; (b) increased costs because of the need to work around existing utility facilities that had not been relocated; (c) crew delays while waiting on decisions regarding unknown utility facilities; and (d) having to schedule work during more expensive seasons, or push the overall construction schedule into the next construction season.

Practitioner Survey

The research team conducted a national survey to identify risk factors affecting the utility process during the project delivery process, primarily during construction. A total of 194 responses included 192 responses from 44 states and two responses from Canada. Respondents included representatives of project owner agencies, consultants, contractors, and utility owners.

The survey instrument included a list of 61 risk factors. Respondents were asked to rate risk factors on a scale from 1 (least frequent) to 5 (most frequent) in terms of how frequently the risk factors contribute to project delivery delays, project cost increases, and utility relocation delays. Overall, the practitioner survey validated the findings of the literature review but also characterized risk factors at a much finer level of disaggregation.

Change Order Analysis

The research team received change order and claim databases from six state DOTs. In total, the research team processed over 150,000 change order and claim records. The research team classified individual change order records as utility-related (UR) or non-utility-related (NUR). To classify change orders, the research team first used commonly used the UR terms and then reviewed the description and justification columns of each change order to assess whether the change order was UR or NUR. The research team also used artificial intelligence (AI) models to detect trends and patterns that could show a change order was UR or NUR. After using the AI models, the research team reviewed individual records to confirm or, if necessary, change the label the AI tools predicted. The total number of UR change orders for the six cases was 11,803.

The research team classified UR change orders according to a list of nine disaggregated reasons. Results were as follows:

• Errors and omissions in PS&E (33 percent).
• Inaccurate or incomplete data about existing or relocated utility facilities (23 percent).
• Changes initiated by project owner, contractor, or utility owner (12 percent).
• Delays getting utility owners to schedule utility relocations (11 percent).
• DSCs (4 percent).
• Difficult or inadequate constructability of highway work or utility relocation (4 percent).
• Inaccurate or deficient utility relocation work (2 percent).
• Delays acquiring or clearing right-of-way or utility relocation sites (2 percent).
• Other (9 percent).

For departments of transportations (DOT) where the change order description was sufficient, the number of UR change orders the research team classified as DSCs was low. For those states, the research team could determine the actual reason behind the change order (even if the DOT had originally classified the change order as a DSC). This result is significant because it could point to many cases in which a change order might be classified as a DSC for convenience or because the official in charge did not have more meaningful categories to choose from, but the actual reason was completely different.

As mentioned, the research team used AI models to detect trends and patterns that could show a change order was UR or NUR. The research team used the change order database from Case 9, which included 102,302 records that had unique entries. Of this total, 95,290 (93 percent) were NUR records and 7012 (7 percent) were UR records. The research team used three vectorization techniques to transform text into numerical data: CountVectorizer, TF-IDF, and bidirectional encoder representations from transformers (BERT). The research team used the vectorized training datasets to train six AI models, as follows: Logistic regression, kNN, multi-layer perceptron classifier, support vector machine (SVM), random forest, and deep learning.

The average classification accuracy of validation datasets for UR change orders ranged from 52–88 percent. Deep learning with the BERT vectorization technique achieved an overall accuracy or 88 percent for UR change orders, followed by random forest with the TF-IDF vectorization technique, which achieved 81 percent.

CASE STUDIES

The research team completed three case studies to highlight exemplary practices on how to manage utility issues, particularly during the construction phase. The three case studies were located in Colorado, Texas, and Virginia.

The Colorado project involved replacing and realigning the existing bridge on SH 82 over a railroad track, the Colorado River, and I 70 in Glenwood Springs. A new pedestrian bridge was designed to carry the existing utility facilities that were attached to the old vehicular bridge. The utility relocation design involved converging all existing lines into a vault, and then elbowing up to the underside of the pedestrian bridge. The Colorado Department of Transportation (CDOT) implemented a utility engineering-based program that included early utility investigations during project deliveries, early identification and resolution of conflicts, and frequent coordination with utility owners. The project had few UR issues during construction. The project had a construction management consulting contract that included utility inspections. The project included the preparation of utility as-builts.

The Texas project involves widening US 281 in San Antonio from a four-lane median-divided cross section to a six-lane freeway with two-lane directional frontage roads. The project is nearing completion. The project is part of the utility conflict management (UCM) implementation at the Texas Department of Transportation (TxDOT). The project also has a construction engineering and inspection (CEI) contract that includes utility coordination and utility inspection services. Most utility owners took care of their own relocations, whether reimbursable or not. Some utility relocations were included in the highway contract. For non-joint-bid relocations, the CEI consultant highlighted the need for extra coordination with utility owners because the contractor wanted to work on several fronts simultaneously. For joint-bid relocations, the construction manager found it efficient to coordinate with the utility contractor directly. Overall, the change order numbers show that the UCM implementation, adding utility coordination and inspection to the scope of the CEI contracts, and other strategies (such as using a right-of-way clearing contract) had a positive impact on the management of utility issues both prior to letting and during construction.

The Virginia project was a 0.8 km (0.5-mi) project in Hillsboro that consisted of two roundabouts, raised crosswalks, sidewalks, a new municipal drinking water system, wastewater treatment facility, stormwater collection system, undergrounding all overhead utility lines, and dark-sky-compliant streetlamps. One third of the project cost went to utility and stormwater infrastructure. The project included strategies to build utility systems in a narrow roadway, utility coordination for relocation work, and close coordination for consecutive and concurrent relocation work. The Virginia Department of Transportation (VDOT) normally relocates utility facilities prior to letting. In this project, utility facilities were relocated during construction to reduce impacts. The only UR change orders were related to the amount of concrete in the duct banks.

FUNCTIONAL REQUIREMENTS FOR A DECISION SUPPORT SYSTEM

The research team prepared a list of requirements for the development of an intelligent decision support system (IDSS) focusing on the classification of UR change orders and identification of

their causes. The requirements include (a) recommendations to improve the clarity, completeness, and conciseness of change order descriptions as new change orders are generated and (b) IDSS components and mockup interface components that illustrate potential workflows.

The recommendations to improve the clarity, completeness, and conciseness of new change orders be included as part of a Help subsystem in the IDSS, as a separate guide document, or as part of an existing construction management software the DOT already uses. For the IDSS components, the research team assumed the IDSS would be installed on a cloud server and that user access to the system would be via a web browser. The cloud server could be owned by the DOT or hosted on a commercial platform. Components include a web application programming interface (API) to extract change order data from an existing project management system (PMS); IDSS components to process and analyze data, generate reports, and manage the system; and a user interface to interact with and run the IDSS.

PROCEDURES FOR CONDUCTING UTILITY INSPECTIONS

The research team developed utility inspection procedures taking into account data collection equipment, software, and protocols. The research team conducted a review of data collection equipment that was suitable for conducting utility inspections. The focus was low-cost devices that could still provide cm-level positional accuracy levels. The research team reviewed unmanned aircraft systems (UASs), smartphones and tablets, and external global navigation satellite system (GNSS) antennas.

Most UAS applications used for inspections involve the use of small rotary platforms. Real-time kinematics (RTK) support is desirable but not essential if ground control points (GCPs) are used in the field. Recent smartphones and tablets have the capability to receive data from multiple GNSS constellations. A wide range of mobile devices are suitable for conducting utility inspections.

The research team also reviewed external GNSS antennas. Of interest here are devices and companion services that offer cm-level positional accuracy at lower costs than traditional GNSS equipment. A typical business model is one in which the cost of the GNSS antenna is low (say $500–$5,000). The receiver provides a positional accuracy between 60 cm (2 ft) and 1.5 m (5 ft) in autonomous mode, but when connected to an RTK correction subscription service, the positional accuracy improves up to 1–3 cm horizontally. RTK subscription rates range from $4,000 per year to $400 per month or $100 per day. Depending on the brand and model, GNSS receivers can connect to public RTK networks for free, but in other cases, users first must pay an unlocking or access fee to the GNSS vendor.

The research team conducted a review of several data collection apps for mobile devices. Of interest were apps that enable users to complete activities such as, but not limited to collecting data using preestablished feature classes and drop-down lists; comparing planned versus as-built locations; gathering unstructured point, line, and polygon data; collecting picture sets and light detection and ranging (LiDAR) data to produce 3D models; and adding comments.

The research team conducted benchmark tests to assess the positional accuracy of the GNSS antennas. On autonomous mode, the horizontal positional accuracy varied from 1–2 m (3–7 ft)

and the vertical accuracy varied from 0.2–9 m. When using RTK, the horizontal positional accuracy varied from 1–4 cm and the vertical positional accuracy varied from 1–10 cm. These results show that low-cost GNSS antennas connected to an RTK network can provide cm-level positional accuracies, which are sufficient for most utility inspection activities. The review confirmed the availability of several apps for mobile devices, which have stakeout functions that enable users to compare design locations versus actual locations on the ground.

The research team identified five data collection cases that apply to a wide range of utility inspection activities that involve verification of locations, areas, and volumes, as follows:

• Project survey control point verification: The inspector occupies one or more project survey control points (SCPs) to make sure the coordinate system parameters used for the data collection are consistent with those used for project survey control.
• Point feature: The inspector uses the stakeout tool of the data collection app to find the point feature and verify whether its location is within a pre-specified tolerance.
• Line feature: The inspector uses the line stakeout tool of the data collection app to find the line feature and verify whether its location is within a pre-specified tolerance.
• Polygon features: The inspector uses the stakeout tool of the data collection app to find each of the corners of the polygon feature and verify whether its location is within a pre-specified tolerance.
• 3D objects: The inspector uses a UAS or a smartphone to capture multiple images around the area of interest. Some devices also have LiDAR data collection capabilities. After processing the data using photogrammetry software, the result is a georeferenced 3D model of the feature and area of interest.

RECOMMENDATIONS

Recommendations Prior to Letting

Conduct utility investigations systematically.

A recommended practice is to conduct utility investigations as early as possible during project delivery, with each quality level contributing to a reduction in the level of uncertainty about utility facility locations depending on project needs. General guidelines are as follows:

• Preliminary design (prior or up to 30 percent design): Conduct preliminary utility investigation based on existing records (i.e., quality level D [QLD]), conduct utility investigation using geophysical techniques (i.e., quality level B [QLB]), and conduct utility investigation based on aboveground utility facilities (i.e., quality level C [QLC]). In general, it is advisable to first gather QLD data for the entire project, and then schedule the collection of QLB and QLC data. Because of the cost of QLB data, it is best to use geophysical techniques wherever there is a need for reliability in the horizontal location of utility facilities. Often, this requirement makes it necessary to use QLB for the entire project, but in other cases, it is possible to limit the collection of QLB to strategic areas.
• Detailed design (30–60 percent design): Conduct utility investigation using test holes (i.e., quality level A [QLA]). Because of the cost of test holes, it is best to strategize test hole locations using criteria such as outcomes of earlier utility investigation activities and the identification of locations where knowing the elevation of an underground utility

• facility is essential (e.g., gas or high-pressure pipeline crossings, structure foundations, and culvert inlets and outlets).

Apply a UCM approach to identify and resolve utility conflicts.

UCM is a comprehensive multi-stage process that involves the systematic identification and resolution of utility conflicts. UCM stages can vary depending on project characteristics. A recommended practice for UCM is to depict the location of utility conflicts on a utility layout and use a utility conflict list (also called a utility conflict matrix) to document each conflict, the process to analyze resolution alternatives, and the alternative that was finally selected. Specific recommendations to apply UCM effectively are as follows:

• Involve all stakeholders in the UCM process. UCM is a team effort that involves all stakeholders, not just the utility coordinator. The level of involvement depends on the role of each actor.
• Document each conflict using the utility layout, utility conflict list, and companion documentation (e.g., project files, pictures, specifications, schedules, right-of-way acquisition plans, and drainage design files).
• Identify and analyze conflicts in preparation for the completion of milestone deliverables. This UCM approach is proactive, as opposed to first preparing milestone deliverables (e.g., at 30-percent design, 60-percent design, 90-percent design, or 100-percent design) and then conducting the utility conflict analysis.
• Use content from the UCM process to prepare the utility statement that is necessary to prepare the construction bid package, showing utility work completed prior to construction, utilities not in conflict with the project, and utility work that must be completed during construction.

Conduct constructability reviews whenever utility facilities are involved.

It is common to conduct constructability reviews in situations where highway design features affect existing utility facilities (provided a proper utility investigation reveals the location and impact associated with these facilities). Constructability reviews of utility relocations are much less common. Having a construction engineer review utility conflicts and utility relocation plans helps with the identification of issues that utility relocations might face in the field as well as issues the highway contractor might find during construction. Effective constructability reviews often involve utility owners.

Include utility relocations in assessment of critical path for the project.

Often, utility relocation schedules only consist of a highly aggregated list of tasks and durations, missing important information to put utility relocation activities in proper context with respect to the highway construction project. Effective utility relocation schedules are those that are organized into manageable, logical phases, and include activities, durations, and milestones. Commonly used project management software should be used to prepare Gantt chart schedules that include these elements and enable the identification of schedule dependencies and critical paths. Ensuring that utility relocation schedules are as accurate as possible is important because, ultimately, if a utility owner does not clear its utilities on time in an area where the highway contractor needs to work, the DOT can be liable for delay costs.

Prepare robust utility relocation documentation.

Required supporting documents to prepare utility agreements are the utility relocation plans, the utility relocation schedule, and the utility relocation cost estimate. Consolidated plans and schedules for all utility relocations are also required for inclusion in the construction bid package. The research team prepared a checklist that includes both required elements and enhancement elements for inclusion in utility agreements and the construction bid package. Required elements are those that must be included in a required document or deliverable. Enhancement elements increase the completeness and quality of documents or deliverables that already include the required elements.

Develop a utility construction plan and include it in the highway contract.

A utility construction plan (UCP) assembles elements from the utility relocation plans and utility relocation schedules into one document to create a narrative that explains how the highway construction can be affected and specific steps to manage those impacts. UCPs focus primarily on utility relocations that are not included in the highway contract under the assumption that the construction bid package already includes all the necessary information for in-contract utility relocations. Utility coordinators should begin developing UCPs well in advance of the letting date when it is clear which utility relocations will not be cleared prior to letting.

Use right-of-way clearing contracts for utility relocations.

Right-of-way clearing contracts outside the highway contract are useful for clearing the right-of-way in preparation for utility relocations, particularly in heavily vegetated areas. In a typical situation, only one right-of-way clearing contract is necessary to prepare the area for all utility relocations, therefore reducing the risk of overpaying for multiple right-of-way clearing activities. Another benefit is to increase the chances each utility owner will relocate correctly and on schedule. In addition, it may be possible to use right-of-way clearing contracts to remove abandoned lines when the owner cannot be located or is out of business.

Recommendations During Construction

Stake right-of-way and maintain markers for utility relocations.

When a utility owner places facilities on a project without knowing with certainty where the right-of-way line, the risk exists that utility crews will guess at the right-of-way line and place the utility facility in the wrong location or, worse, on private property. Having to relocate utility facilities a second time to correct the error can affect the sequence of highway construction. Staking the right-of-way before utility relocations start would help to prevent this issue from occurring. Staking the right-of-way is particularly critical on roadways where the DOT did not acquire right-of-way for the project, and many monuments have been knocked out over the years as a result of roadside maintenance operations, fence construction, and utility installations close to the right-of-way line. However, staking the right-of-way on proposed right-of-way is also beneficial, particularly if the new corners have not been set or contractors knocked them out placing new fences.

Develop a common repository of utility data and other project-related data.

One of the most time-consuming tasks during construction is to keep all stakeholders on the same project datum and checking for field locations not matching utility plans or project files. Using an incorrect datum, scale factor, or benchmarks is a common issue that requires multiple meetings to resolve the differences and then having to revise the plans. An effective strategy to address these issues is to set up a cloud-based shared drive to store up-to-date information that all stakeholders can

access. Having a common set of benchmarks and associated metadata and making this information available ensures that all stakeholders use the same datum for all field measurements, including utility relocations and productions of utility as-builts. The shared drive can also store a list of all contacts including utility owners, contractor personnel, inspectors, and emergency numbers, as well as copies of project files and as-built utility plans as they become available.

Schedule utility preconstruction meeting or include utility owners in highway preconstruction meetings.

Preconstruction meetings are standard highway construction events. These meetings set the stage for the establishment of communication protocols and other procedures among stakeholders, including contractor, subcontractors, DOT construction manager, design consultants, surveyor, inspectors, and others. For small projects or projects that do not have complex utility relocation issues, including utility owners in the highway preconstruction meeting is certainly advisable. For large highway construction projects, or in situations that involve complex utility issues to address during construction, it is best to schedule a separate utility preconstruction meeting.

Schedule recurrent utility coordination meetings during construction.

Utility owners that are relocating facilities during highway construction should have frequent meetings with the highway contractor and other DOT representatives (e.g., construction manager, inspector, and surveyor). A recommended practice to set up recurrent meetings at the job site office (e.g., weekly). As needed, meetings could also take place at specific locations on the project to assess field conditions. Examples of items to discuss include relocation status and schedules, coordination and cooperation needs; traffic control; environmental compliance; Build America, Buy America (BABA) Act requirements; and when materials will be on site for inspection.

Use plastic pipe to mark underground lines.

One of the challenges when installing new underground utility facilities or when exposing existing lines is once the excavation is backfilled, it is quite difficult to remember where the facility was located. Even when accurate X-Y-Z data are collected, construction crews do not necessarily have ready access to the data. Placing a 5-cm (2-in) plastic pipe vertically on top of the line when the trench or test hole is still open and allowing the pipe to protrude slightly above the ground level enables all stakeholders to easily locate the underground facility. The contractor, utility owner representatives, or a surveyor can also use a tape to verify the depth of the line relative to any work that may be happening on the surface. This low-cost technique is particularly effective in situations where it is not clear whether all stakeholders using the same datum.

Use utility layout to show abandoned utility facilities.

If a contractor finds a utility facility that was not included in the utility plans or listed in the utility conflict matrix, it is impossible to know without more information whether the line is active, inactive, out of service (temporarily or permanently), or abandoned. To minimize its liability, the contractor stops working in the area until a positive confirmation arrives about the status of the line. A strategy to manage abandoned utility facilities during construction is to prepare plan sheets that show all abandoned lines that are found within the project limits, along with information about the owner and the agreed upon disposition of the line (e.g., removal or cut and fill with grout).

Conduct utility relocation inspections systematically.

Requirements for the inspection of utility construction vary with the complexity and location of the utility work and the associated impacts on the transportation facility. For smaller installations, it may be sufficient to spot check for general conditions of the relocation, traffic control, and safety. In other cases, the complexity of the utility work may require continuous and close observations of (a) construction methods, including excavation, installation, backfilling, and restoration, and (b) alignment and dimensions (i.e., X-Y-Z coordinates) of the utility facilities within the right-of-way. In addition to verifying actual locations, an important focus of the inspection job is to verify the utility facilities as installed are not in conflict with adjacent facilities and structures. Even if a facility is installed properly according to the plans, checking for conflicts in the field helps to uncover hidden situations the contractor or other utility owners might have missed otherwise, but which need to address, sometimes as soon as possible, to avoid the risk of project delays.

Improve the quality of change order documentation to facilitate future analyses.

Extracting data about UR change orders and claims from the DOT’s data repository is essential for understanding how UR reasons can cause issues during construction. One of the strategies to achieve this goal is by improving the clarity, completeness, and conciseness of new UR change orders and claims. Another strategy is to facilitate the extraction of UR change orders and claims from an existing database by implementing a cloud-based decision support system that interacts with the database and includes components that enable the classification of records, various analyses, and preparation of reports. The decision support system could be owned by the DOT or hosted on a commercial platform.

Processing and analyzing change order data in the decision support system would involve a combination of automated record classification and manual review and editing. The research showed the feasibility of using AI models to automate the detection of UR change order records. The research also showed the feasibility of using trained AI models using data from one DOT to classify change order records from other DOTs. The tests showed that if the change order description structure is similar to that used for training AI models, using AI models to classify change order records from other DOTs is feasible.

SUGGESTED RESEARCH

Effective classification of change orders to facilitate future analyses.

DOTs organize and group change orders in many different ways. Typical categories include contract administration, errors and omissions in PS&E, DSCs, change in scope, right-of-way, and utilities. The level of disaggregation in change order classifications varies widely. NCHRP 15-69 focused on UR change orders. To assist with the change order analysis, the research team successfully used AI models to detect UR change orders. Research extending the use of AI models to all other reasons that cause change orders and claims, coupled with a judicious manual review of a significant sample of change order records, would help DOTs in developing a much more accurate understanding of what is causing change orders. The research would also produce an updated classification of change order reason codes and recommendations to improve change order descriptions.

Building information modeling (BIM) architecture for utility facilities.

DOTs are quickly adopting BIM to develop and deliver highway projects. FHWA is also heavily promoting BIM

for infrastructure as a collaborative work method for structuring, managing, and using data and information throughout the lifecycle of assets withing the right-of-way. Research is needed to develop and test a BIM architecture for utility facilities in ways that facilitate (a) data exchange during all phases of project delivery without losing data integrity and completeness and (b) integration with all other aspects of BIM-based project delivery activities, from design to construction and post-construction. Developing and testing the BIM architecture for utility facilities must comply with existing and emerging data exchange industry standards.