Time-Related Incentive and Disincentive Provisions in Highway Construction Contracts (2010)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

10 Cost It is assumed that acceleration associated with I/Ds increases project cost. To what degree costs are increased is a function of unique project features and the level of acceleration requested by the STA. The majority of responses to the Phase I e-mail interview form indicate that the STAs perceive the increased cost to be 10% or less. Phase I in-depth ranking forms concur with this perception (Figure 4). On average, interviewees felt that the impact of I/D provisions on costs was neutral. A comparison of A+B projects and non-I/D projects of similar scope in Minnesota showed that the A+B projects had an initial bid price increase of 7.5% when compared with similar non-I/D projects (16). Market Influences Phase II research results revealed that contractors agreed that under the low bid system used by STAs, influences from market conditions (number of bidders, backlog, and avail- ability of future work) are a primary factor in determining the cost associated with the use of I/D provisions (Figure 5). Understanding that market conditions influence bid prices as much as or more than acceleration costs has an impact on the effectiveness of I/D provisions. The need for a competitive market environment is especially important for the effec- tive use of I/D provisions. One contractor’s comment from the Phase II research effectively captures this point: “The STA must make the decision about the level of importance or priority for early or accelerated completion of a project, then let the market decide what that priority is worth.” Further evidence that market influence can trump accelera- tion costs comes from two projects bid by the Kansas Depart- ment of Transportation (Kansas DOT). These two projects required the bidders to provide an alternate lump sum bid price for accelerating the projects. The base bid for the projects was based on completing the projects in two construction seasons. The additive alternate bid item was designed to capture and compensate the contractor for the acceleration costs associated with completing the projects in one construction season. In both cases, the alternate bid item for acceleration costs did not change the order of the bidders nor does it appear that Kansas DOT was charged a premium for accelerating the projects (Tables 4 and 5). There is a valid argument that the contractors included an acceleration premium in their base bid because they anticipated Kansas DOT’s desire to complete the project in one season and that they did not want to expose their estimated acceleration costs to other competitors. It is the research team’s informed opinion that if any of the acceleration costs required to complete these projects in one season was passed on to Kansas DOT, it was minimal. This conclusion is based on the fact that one member of the research team was formerly employed as the chief estimator for one of the unsuccessful bidders on the Harvey County project, and the base bid prices reflected con- tractor bid prices that were very competitive (acceleration costs were offset by lower profit margins). It should be noted that both projects were completed within one construction season. These two Kansas DOT examples are presented as an illus- tration of how market influences often trump acceleration costs. Note that the difference between the low bidder and the average bid on the first project was 12% and was 16% on the second example. Because these projects required the contractors to provide a lump sum bid for acceleration costs, they provide a unique picture of the contractors’ approach to acceleration costs that are not visible in the bidding process for other types of I/D projects. In the research team’s experi- ence estimating and bidding on over 40 I/D projects, market conditions are the primary influence on the final cost paid by STAs. This applies to all types of I/D projects. STAs must understand that for a given I/D project the cost that they ultimately pay for acceleration and the level of acceleration that they receive is highly dependent upon the market conditions at the time the project is bid. Not only are acceleration costs C H A P T E R 2 Discussion of I/D Impacts on Project Factors

11 1.0 0.5 1.7 1.0 –0.8 –0.7 –1.1 –0.2 –1.0 –1.3 –3.0 –3.0 –3.0 –2.0 –3.1 –5.0 –4.0 –3.0 –2.0 –1.0 0.0 1.0 2.0 3.0 4.0 5.0 all agency admin. agency field contractor admin. contractor field In what manner and to what degree do time-related I/D provisions impact PROJECT COST? upper quartile average lower quartile significant beneficial impact considerable beneficial impact moderate beneficial impact slight beneficial impact neutral significant detrimental impact considerable detrimental impact moderate detrimental impact slight detrimental impact Figure 4. Agency and contractor Phase I on-site interview ranking of I/D impacts on project cost. 1 2 3 4 5 6 Le ve l o f A gr ee m en t Contractor Responses Contractor Perspective Median (All Contractors) strongly agree agree strongly disagree disagree not sure Figure 5. Contractor perspectives on market influences and cost. Bidder Base Bid: Complete in Two Seasons Alternate Bid: Acceleration Cost to Complete in One Season Total Bid to Complete in One Season Bidder 1 $18,482,549 $1 $18,482,550 Bidder 2 $20,636,141 $812,270 $21,448,411 Bidder 3 $21,035,433 $1,202,000 $22,237,433 Table 4. Kansas DOT Harvey County accelerated bid results (I-135, October 1999).

12 Bidder Base Bid: Complete in Two Seasons Alternate Bid: Acceleration Cost to Complete in One Season Total Bid to Complete in One Season Bidder 1 $16,279,056 $0 $16,279,056 Bidder 2 $16,877,890 $5,000 $16,882,890 Bidder 3 $17,962,028 $1,000,000 $18,962,028 Bidder 4 $19,644,504 $1,250,000 $20,894,504 Bidder 5 $20,945,708 $472,271 $21,417,979 Table 5. Kansas DOT Sedgwick County accelerated bid results (US-54, January 2003). 1 2 3 4 5 6 Le ve l o f A gr ee m en t Contractor Responses Contractor Perspective Median (All Contractors) strongly agree agree strongly disagree disagree not sure Figure 6. Phase II contractor feedback about acceleration costs and project features. unique to each project, they are unique to each contractor as well (Figure 6). The benefit for the STA of competition in the bidding process cannot be over stressed. Cost Components Even though market factors are the primary influence on costs associated with the acceleration required by I/D provi- sions, it is useful for a STA to have a basic understanding of construction cost components and how they are impacted by acceleration. A typical unit price found in a contract for constructing a highway is the sum of four individual unit cost components: materials, labor, equipment, and overhead. Profit is added to the sum of these components. Table 6 contains a simplified list of cost components; from a contractor’s per- spective, these macro-components are further subdivided into multiple sub-components. Table 6 provides a brief description of acceleration impacts on each of the cost components. Acceleration impacts on cost will also vary by the type of work. For example, a unit of structural concrete will be impacted to a greater degree than a unit of paving, because the structural concrete has a larger labor component per unit than paving (Figure 7). Certain aspects of acceleration associated with the use of time-related I/Ds may have minimal impact or reduce the actual cost of construction. Examples of this include the following: • Improved management and scheduling can enhance resource use. • Innovative methods and materials used to meet the schedule demands of an I/D provision may result in cost savings. • When contractor resources and project conditions are such that multiple crews can work on an item simultaneously without interfering with each other (e.g., multiple cut/fill locations), cost impacts are not an issue. However, highly phased projects do not offer the same opportunity to add resources on the same work item. Extend- ing hours and work weeks may be the only way to accelerate construction because of limited work areas. Thus, acceleration costs will be greater when the construction sequence and plans have limited areas of simultaneous work activities. Contractors are acutely aware of their unit costs, and the vast majority of successful contractors go to great efforts to track them. Contractors’ bid prices rely heavily on past experience (historical costs); even with this detailed information, esti- mating the impact of acceleration on unit costs is as much art as it is science. This is one of the inherent risks that contractors face each day. Over time as more I/D projects are contracted,

a STA may develop a historical database that can assist in estimating the cost impacts of the use of I/D provisions. From a STA’s perspective, estimating the cost impact of acceleration should be based on a macro level. An overall cost adjustment factor should be applied that considers the estimated net effect of market influence and acceleration cost impacts. A primary concern of STAs is understanding the relationship between I/D amounts, cost, and acceleration. For a STA, there is an optimum minimum I/D amount that will motivate a contractor to safely complete a highway construction project within the time frame set by the STA. In other words, a STA would like to know: what is the minimum I/D amount that will ensure that schedule goals are met? The answer to this question is complex and in almost every case unquantifiable. It is impractical for a STA to pursue models or algorithms that attempt to estimate the minimum 13 Cost Component Discussion of Acceleration Impacts Materials Typically more affected by supply and demand issues (market factors) than by acceleration. Projects that require a sustained high level of acceleration are an exception to this general rule. Some suppliers may not be able to meet the accelerated supply schedule, thus reducing competitive forces on material pricing. Labor Typically the most affected cost component. Acceleration directly impacts labor rates through overtime and premium pay. Production rates are typically lower during periods of extended working hours and for multiple shift scenarios. Equipment Production rates are typically lower during periods of extended working hours and for multiple shift scenarios, thus increasing unit costs. Additional equipment resources (lease/rent) may be required to meet accelerated schedule requirements resulting in higher costs. Overhead Home office overhead is a function of annual revenues. Assuming a contractor is not forced to forego other revenue opportunities due to the resource requirements of an accelerated project, home office overhead would be reduced. Project overhead rates are typically increased for an I/D project due to the need for additional management resources. Profit Market driven. Table 6. Description of acceleration impacts on cost components. 0% 20% 40% 60% 80% 100% 120% structural concrete (normal) structural concrete (accelerated) concrete pavement (normal) concrete pavement (accelerated) overhead equipment labor materials 111% 108% Figure 7. Example of acceleration cost impact.

14 I/D rate that will affect acceleration. The primary factor that makes such models/algorithms impractical is the impact of market influences on contractor bid prices and scheduling. It is sufficient to state that market factors can have a greater influence on the final cost paid by a STA than any other factor. That is to say that the competitive bid environment can trump cost impacts created by I/D provisions (almost always to the STA’s advantage). This does not mean that STAs should not concern themselves with optimizing the I/D rate. Rather, it emphasizes the necessity of STAs to gain a more thorough understanding of the impacts of I/D provisions and the inter- dependent nature of many I/D factors. One comment was received from contractors related directly to this issue of what minimum I/D rate will motivate acceleration: “Five to seven percent of the contract value should be the floor for I/D. This is the amount that it takes to get the contractor’s attention and makes the endeavor worthwhile.” Targeting the maximum incentive at a given percentage of the total contract value may be desirable from a policy standpoint, but will fail to yield effective results from I/D provisions. Incentives and disincentives should be adjusted based on the project conditions, market influences, and the level of acceleration required to meet schedule objectives. Other Cost Considerations Incentives Serve as Alternative Revenue for Contractors The low bid system used by STAs creates a highly compet- itive bidding environment. Rarely does construction demand stretch the resources available in the contracting industry. As a result, competition dictates that the majority of I/Ds are passed through to the STA. This practice gives the best qualified contractor for a specific project an advantage at the bidding table. Anticipated incentive earnings are treated like contract revenues; which mean I/D provisions create acceleration at minimal cost to the STA. In effect, when a contractor is con- fident that his/her available resources will meet the objective of an I/D provision and an incentive will be earned, the bid can be reduced by the amount of the estimated incentive. Thus, incentives are treated like regular contract revenues that replace portions of contract profits (in some cases, cost as well). This does not negate the earlier discussion of market factors, but draws focus to the fact that markets are predominantly more competitive than they are flush with available work. Contractors that responded during Phase II were unanimous in their agree- ment that this practice is not only prevalent, but that it is advantageous to the STA and to the contractor. The compet- itive low bid system used by STAs results in 75% or more of I/Ds being incorporated into the bid. This practice provides an advantage to the best qualified contractor while achieving the STA’s schedule goals at minimal cost (Figure 8). Unbalanced Bids The practice of adjusting unit priced bid items for anticipated underruns and overruns (unbalanced bids) is amplified by the use of I/D provisions. In the case of an A+B contract where the contractor determines the contract time, it is common practice for contractors to move some amount of money from the B portion (time) to the A portion (traditional bid items). Consider the example shown in Figure 9. This same scenario holds true for an I/D project where time is determined by the STA as well. When the incentive is bid into the project (passed on to the STA because of market factors) unit prices must be adjusted to affect this bid reduction. At a minimum, the prudent contractor will reduce items that are anticipated to underrun. It is likely that the contractor will arrive at the final bid (reduced total/unbalanced bid) by adjust- ing both underrun and overrun items. While not unanimous, the Phase II research results agree with the research team’s experience that unbalanced bids are amplified by the use of I/D provisions (Figure 10). This practice of unbalancing bids is not wrong or unethical unless it is carried to the extreme, which may lead to dis- qualification of the bid. Unbalancing is inherent to unit price 1 2 3 4 5 6 Le ve l o f A gr ee m en t Contractor Responses Contractor Perspective Median (All Contractors) strongly agree agree strongly disagree disagree not sure Figure 8. I/D treated as alternative contract revenue.

15 Balanced Bid Unbalanced Bid A Subtotal = $50,000,000 * Subtotal = $51,000,000 *** B Subtotal = $10,000,000 B Subtotal = $9,000,000 *where Unclassified Excavation = $2,500,000 (500,000 CY @ $5.00) ***where Unclassified Excavation = $3,500,000 (500,000 CY @ $7.00) Total Bid = $60,000,000 Total Bid = $60,000,000 Final Contrac t Amount Paid by STA = $50,500,000 ** Final Contrac t Amount Paid by STA = $51,700,000 **** **Unclassified Excavation overrun = $500,000 (100,000 CY @ $5.00) ****Unclassified Excavation overrun = $700,000 (100,000 CY @ $7.00) STA Cost Increase from Unbalanced Bid = $1,200,000 A Figure 9. Unbalanced bid example. 1 2 3 4 5 6 Le ve l o f A gr ee m en t Contractor Responses Contractor Perspective Median (All Contractors) strongly agree agree strongly disagree disagree not sure Figure 10. Contractor feedback about unbalanced bid practices on I/D projects. contracts and has been an accepted buyer’s risk borne by STAs for decades. STAs should be aware that I/D provisions amplify this practice, as one contractor stated in the feedback to Phase II research: “Contractors often perceive STAs as being hesitant to pay the incentives the STA has contracted to pay in contracts containing significant I/D provisions. While the STA may have made the decision that there is value in paying these incentives, Contractor experiences often indicate this commitment breaks down when the due date for paying these incentive approaches. As a result, Contractors often budget this incentive in the cost of their products; meaning the STA is paying the incentive with each product delivered and the contractor’s scheduled completion date may actually be later than the STA realizes. Negotiating to end days of damages versus starting days of incentive favors the Contractor most often.” Another contractor expressed the following: “Typically, only the savvy Contractor can/will un- balance a bid. Is that a bad thing; I think not.” Thus, STAs should take extra care when estimating plan quantities on projects that will include a time-related I/D provision. The degree to which unbalancing occurs as a result of I/D provisions is related to how the I/D provision addresses adjustments to time during execution of the contract (no-excuse or day-for-day) and the extent to which time impacts are antic- ipated by the contractor. In simplest terms, if the contractor anticipates that negotiable time delays will occur, they will be more likely to unbalance a bid by reducing the B portion or anticipating incentive payments. Risk Allocation Between Contract Parties I/D provisions normally change the way risks are allo- cated between contract parties. Most contractors are capa- ble of estimating their costs and their schedule. When I/D provisions shift risk for certain time impacts (weather, util- ities, quantities, etc.) to the contractor, the cost to the STA increases. This increase in cost to the STA is directly related to two factors: (1) Is the risk quantifiable (how well can the contractor estimate the impact)? and (2) Is the risk beyond the control of the contractor (can the contractor do any- thing to mitigate the risk)? An example of changing risk allocation is a no-excuse clause which requires the contractor to complete the project within a given time frame regardless of impacts that are encountered on a project. Without debating whether these clauses are fair or not, it is safe to say that they impact the final cost paid by the STA. In the case of a no-excuse clause, the contractor has

16 assumed all risk associated with weather, site conditions, and so forth. There is no accurate way to estimate the schedule impact of these types of risks. In most cases like this, contractors will be more reluctant to pass any potential incentive earnings back to the STA, because the potential incentive earnings are so dependent on factors that cannot be accurately quantified. Contractors responding to the Phase II research agreed that STAs will incur higher costs in the long term as more risks are shifted to the contractor (Figure 11). One contractor offered the following comment regarding the issue of risk allocation and long-term costs to the STA: “Yes, if the contract provision begins to look too much like a gamble to the Contractor, the Contractor will acknowledge the potential damages by budgeting to incur these. The likelihood that the STA then pays for the potential damages, but then never recoups the damages tends to favor the Contractor.” The manner in which a contractor estimates the impacts of these types of factors is unique to each contractor. One common thread between contractors is the tendency to err on the side of caution when faced with estimating factors that are difficult to quantify or are beyond their control. As a result, STAs need to understand that transferring risk to the contrac- tor, especially risks that are beyond the contractors’ control or those that cannot be quantified will increase the STAs costs in the long run. Methods of risk sharing between the STA and contractor are a critical need that should be addressed by all contract parties. Innovation During the Phase I interviews, contractors and STAs agreed that time-related I/D provisions drive innovation. They also agreed that innovation was a positive by-product of I/D pro- visions. However, some of the contractors voiced a concern that STAs were not always open to the proposed innovation or that the STA expected to negotiate time as a part of the approval process for the innovative method. Contracts are changed in many different manners. Each STA has detailed specifications dealing with the prosecution of contract changes. The use of an I/D provision does not normally modify any of these contract change processes. What is changed by the use of an I/D provi- sion though is the value of time. When a STA uses an I/D provision, they are placing an increased emphasis on time. This emphasis is expressed by placing a monetary value on early completion. This value of time is almost always based on RUC. Except in rare cases such as emergency projects, there is a practical limit both to the value of early completion and to the amount that a project can be accelerated. Thus, the practice of capping the maximum amount of early completion incentive that can be earned is warranted and would be considered a best practice for STAs to implement with I/D provisions. If an I/D provision caps the maximum incentive that can be earned for early completion, time savings should be a non-negotiable issue when a contractor proposes an innovative method. By capping the incentive, the STA has limited their potential to overpay for early completion. Under the capped incentive scenario, contractors should not be required to negotiate a new schedule simply because the innovation will save time and they will earn incentive. Requiring contract time reductions when innovative methods or materials are proposed defeats the motivation for innovation and the associated early completion. Quite often the incentive earned is used to offset additional costs that are connected to the innovative method or material. Phase II research shows that contractors are in harmony on this issue, with four of the five contractors strongly agreeing that approval of innovative proposals should be independent of the time savings associated with the innovative proposal (Figure 12). Contractors offered the following comments supporting their perspective on this issue: • “The STA also realizes cost & time savings—their own inspec- tion staff, QA or QC staff, Management overhead and or Consultant inspection fees.” 1 2 3 4 5 6 Le ve l o f A gr ee m en t Contractor Responses Contractor Perspective Median (All Contractors) strongly agree agree strongly disagree disagree not sure Figure 11. Long-term cost impacts of shifting risk to contractors.

• “The STA must be unequivocal within its own management and engineering staff that contractor proposals that save time may result in added profit for the Contractor, and this is the intent of the contract provision—to provide both a financial incentive to the Contractor to minimize the construction duration.” • “It is important for the STAs to understand that each idea or dollar is considered an off-set to each other. Put another way a dollar can only be saved against a specific idea. The next idea is considered a separate negotiation. Money can only be saved one time; against one idea. Agreement to the first idea for money savings does not carry on for the life of the job. It is a 1 to 1 type concept.” This issue should not be viewed as contractor posturing. When dealt with properly by capping the incentive at an amount that will motivate the contractor to meet the STAs desired duration for completion, a win-win scenario is created. The STA and public receive an accelerated project at a price that they pre-determined is desirable to them and the contractor is motivated to pursue innovative methods and materials that will improve the odds of early completion. Contract Administration In contrast to typical highway construction contracts that use liquidated damage clauses, an I/D provision estab- lishes a sizable pool of money for early completion. The adage stating that “time is money” is appropriate. I/Ds def- initely alter the contract administration landscape. When schedule impacts begin to threaten either party’s claim to the pool of money set aside for early completion, the poten- tial for conflict will be elevated. Time-related I/D provi- sions magnify the need for contract administration policies that provide clear direction for determining how contract time is charged and modified throughout the life of the contract. The treatment of contract time charges varies widely across STAs. Climate variation and local statutes are the primary reasons that STAs have different methods for determining how time is measured for highway construction projects. Because of these regional differences, there is not a single optimum solution for time measurement. Implementing I/D provisions may require a STA to modify its standard specification for time measurement (prosecution and progress). The following items highlight the primary areas that should be considered when developing or modifying a specification for measurement of time on an I/D project: • For an I/D contract, the preferred method of measuring time is the calendar day. The use of working days or modified calendar days (6 day week, no holidays, credit for abnormal weather, etc.) can introduce ambiguity into the schedule. • Contractors should be required to submit a baseline schedule for approval. Once approved, this baseline schedule should be used to evaluate all contract changes. • The process for evaluating the impact of excusable delays must be clearly defined. • Schedule impacts should be evaluated and agreed to in a timely manner. Once agreed to, the changes should be reflected in an updated baseline schedule. • The occurrence of excusable delays erodes the effectiveness of I/Ds. Plans and specifications for I/D projects should be carefully reviewed for potential schedule impacts arising from project unknowns and plan errors. The effective use of I/D provisions entails some transfer of risk to the contractor. Otherwise, the STA is left vulnerable to paying an incentive for early completion when the project was not actually completed early. On the other hand, when too much risk is transferred to the contractor, a disincentive may be charged to the contractor when he/she actually completed a project on time or early but encountered delays beyond his/her control. Risk sharing strategies should be implemented that 17 1 2 3 4 5 6 Le ve l o f A gr ee m en t Contractor Responses Contractor Perspective Median (All Contractors) strongly agree agree strongly disagree disagree not sure Figure 12. Contractor agreement on the treatment of contract time coupled with innovative proposals.

18 allow both parties to reasonably meet their objectives. Con- tractors that participated in Phase II believed strongly about this issue (Figure 13). One of the contractors added the following comment about risk sharing and I/D provisions: “The STA must recognize and accept that a Contractor is not going to accept additional risk when negotiating change orders for a contract that contains significant I/D provisions. This becomes acutely problematic when the contract includes a significant damages clause associated with a “due-by” date that cannot be changed. If a change order is needed to add some small, but necessary, product to the contract and negotiations breakdown as to whether or not the duration required for delivering this product results in a project completion after the due date, nobody wins.” Some examples of risk sharing strategies that have been used by STAs were identified through the research project. A brief description of these strategies follows: • In lieu of a strict no-excuse clause where the contractor bears all of the risk for schedule delays, a modified no- excuse clause should be considered. A modified no- excuse clause has been successfully used by FDOT. Under FDOT’s modified no-excuse clause, contract time may be adjusted when an excusable delay has a schedule impact greater than 15% of the remaining contract time when it occurs. This type of solution is a compromise that allows the contractor to potentially recoup some of its acceler- ation costs incurred while overcoming excusable project delays. • The VDOT has used a contractually binding decision making matrix that defines the time frame for making project decisions. A time frame for either making a decision or elevating the issue up the chain of command is specified for the contractor and STA management hierarchy. Similar arrangements are common to partnering agreements. Incorporating such a decision timetable into the contract relieves some of the risk borne by a contractor under an I/D provision. • Time-related overhead (TRO) has been used as a bid item by the California Department of Transportation (Caltrans). By doing so, the contractor establishes the daily overhead rate during the bid process. This rate is used to compensate the contractor for the contract time used and for any excus- able delays that occur. This type of specification applies market pressure to the overhead rates that would otherwise be sole source negotiations between the STA and contractor. Staffing Managing an I/D project presents unique challenges to the STA and to the contractor. Extended work hours and 6- or 7-day work weeks are not uncommon. The in-depth interviews conducted during Phase I with STAs and contrac- tors revealed that both STAs and contractors acknowledge that staff fatigue and burn out occur on I/D projects. But, the in-depth interviews did not produce any results that would point to effective management techniques for dealing with staff fatigue and burn out. The use of consultants for project management and inspection by STAs is becoming more common according to a report by the General Accounting Office (17). Using consultants on an I/D project adds a layer of complexity for the STA and the contractor that does not solve the fatigue/burn-out issue; the burden just shifts to the consultant’s personnel. From the STAs’ perspective, the use of consultants is a reality forced by the changing workforce at STAs and difficulties in attracting and retaining staff with key skills (17). Proper consideration should be given to developing and implementing performance measures and oversight for consultants that will ensure that the STAs’ and public’s interest are being served. When consultants are used, STAs should structure the contract with the consultant in a manner that will equitably compensate the consultant for extended hours as well as protect the STA from changes due to the contractor’s accelerated schedule. Because timely decision making is critical on accelerated proj- ects, the STA needs to be sure there is clear understanding with 1 2 3 4 5 6 Le ve l o f A gr ee m en t Contractor Responses Contractor Perspective Median (All Contractors) strongly agree agree strongly disagree disagree not sure Figure 13. Contractor perspective on risk sharing.

the consultant on decision making authority so that the use of a consultant does not introduce another layer in the decision making process and delay the contractor. One technique that contractors were in favor of was a “flex start.” When STAs allow the contractor to adjust the start of an I/D project into the contractor’s portfolio of work where it fits best, some staffing and resource issues are alleviated (Figure 14). Contractor feedback regarding flex starts was as follows: • “If the STA can offer flexibility in when a project is scheduled, the cost to the STA will be lowered.” • “Consideration should be given of flex starts up to 180 days, not limited to 90 days.” Except when the completion of a project is tied to a critical date (school opening, special event, etc.), flex starts should be allowed whenever possible to allow contractors the best oppor- tunity to balance resources between projects. Doing so will minimize the cost associated with an I/D provision and should contribute to maintaining acceptable quality levels. Given materials that comply with specification, quality is most affected by the human resources used on a project. Many so- lutions exist to add equipment and material resources to a project. However, it still takes people to operate the equip- ment and incorporate the materials; adding human resources while maintaining quality is a much more difficult task. Flex starts are a valuable tool for the contractor to manage staffing issues created by I/D provisions. Quality Maintaining acceptable quality levels for work that is performed under a time-related I/D provision is an issue that should not be overlooked. In-depth interviews with STA and contractor personnel revealed that I/D provisions have no impact on the quality of work. STA personnel were unanimous in their statements that enforcement of speci- fied minimum quality requirements are independent of time- related I/D provisions. Results of the e-mail interview form generally agree with the in-depth Q&A responses, but do indicate that some quality deficiencies are related to I/D provisions (Figure 15). An analysis of 477 projects finalized by a STA for fiscal year 2007 was performed by the research team. No statistically significant difference between the quality of I/D and non-I/D projects was found. A summary of the analyses performed and the conclusions drawn is as follows: • The population of 477 projects finalized for FY2007 included 455 non-I/D projects and 22 I/D projects. • Of the 477 projects, 157 of them had a quality adjustment made to the final contract revenues. • 19% of the non-I/D projects incurred a quality deduction while 18% of the I/D projects incurred a quality deduction. • The net quality adjustment (sum of quality incentive and disincentive) as a percentage of contract value was 0.39% for non-I/D projects and 0.45% for I/D projects, thus the net quality adjustment was higher for I/D projects, indicating slightly higher quality. • An analysis of variance (ANOVA) was used to evaluate the net quality adjustment as a percent of contract value for 19 1 2 3 4 5 6 Le ve l o f A gr ee m en t Contractor Responses Contractor Perspective Median (All Contractors) strongly agree agree strongly disagree disagree not sure Figure 14. Contractor perspective on flex starts. yes, 16% no, 84% yes no Figure 15. Quality deficiencies related to I/D provisions.

non-I/D projects and I/D projects; no statistically significant difference was observed (α = 0.01). • This conclusion regarding no statistical difference between the two data sets (non-I/D and I/D) was unchanged even after removing two contractors from the non-I/D data set that accounted for 20% of the non-I/D projects that were charged only negative quality adjustments. • A third analysis evaluated the results for a single contractor that included 4 I/D projects and 3 non-I/D projects; again, no statistical difference existed for the net quality adjustment as a percent of contract value. It is important to realize that the quality adjustments observed in the data set may or may not be impacted by acceleration. For example, the majority of quality incentives are for pavement smoothness. Without knowing the details of every project, it is impossible to know if the paving activities were actually accelerated by the I/D provision. Finally, it should be noted that I/D provisions have been used extensively by many STAs. By all indications, these STAs intend to maintain their use of I/D provisions. Common sense dictates that if quality was negatively impacted by the use of I/Ds, STAs would have discontinued their use by now. While it is comforting to know that time-related I/Ds have little or no perceived effect on quality, it would be remiss to ignore the potential for quality issues to arise from acceleration of the work. Items that should be noted with respect to quality and I/D provisions include the following: • Conformance with specifications is independent of time- related issues; meeting schedule milestones should not influence a decision to accept failing or marginal work. • Schedule milestones may influence contractors to use alter- native methods and materials that would not otherwise be used on non-I/D projects. This does not imply that minimum specifications will not be met. However, although many contractors perform work that exceeds minimum criteria, the STA may perceive a reduction in quality because the finished product is of a lower quality than what the STA is accustomed to, even though minimum acceptance criteria were met. • The potential to trade off quality for time exists whenever quality pay factor specifications are used in conjunction with time-related I/Ds. • Adequate lighting is a necessity for night work. While quality has not been an issue for the vast majority of I/D projects completed, STAs and contractors should be proactive about dealing with the potential for time/quality tradeoffs. Safety Time-related incentive provisions are normally used for one or more of the following objectives: • Shorten contract period to reduce overall exposure and impact of construction. • Complete for special event or winter condition. • Limit capacity impacts during high traffic flows. The construction contract may contain one or more pro- visions to address the objectives of the STA as discussed in detail the Discussion of I/D Variables section in Chapter 4. Without repeating that discussion, it is important in this section to consider how these strategies may impact safety. Each project is unique and needs to be thought through carefully. Not only does the project itself need to be carefully planned but also one needs to anticipate how the project may impact the transportation network in the area with potential secondary impacts. A traffic demand mitigation strategy is commonly an objective to reduce traffic volumes through a project. But, this may result in unanticipated impacts on the parallel network. For example, when I-15 was shut down for reconstruction prior to Salt Lake City hosting the Olympic Games, an increase in off system accidents was noted (18). In their report titled “User Costs on the I-15 Design-Build Reconstruction,” Martin et al. note: “When the I-15 project began on April 15, 1997, increased travel times, queues, congestion, and accidents became common driving experiences. Public support for the project declined. News media reported that people were concerned about increased accident numbers on streets surrounding I-15 reconstruction areas. On January 7, 1998, a Salt Lake Tribune article titled ‘Communities Seek Help With Traffic Trouble’ stated that ‘30 percent of Interstate 15 traffic [had] poured onto city streets,’ that there was a ‘300 percent jump in automobile accidents,’ and that ‘Police Department overtime expenses [had] jumped 87 percent.’” This is not to infer that the overall safety exposure to the public was greater by using this acceleration strategy, but to illustrate that other safety impacts on off system routes can occur and that the planning for the project needs to anticipate these potential tradeoffs. Martin et al. conclude that the design build approach chosen by the Utah DOT was the safest of the alternatives he analyzed. Further complicating the decisions involved with accelerat- ing work activities is that there are often other tradeoffs involved when minimizing impact to the traveling public through a construction project. Night work is increasingly used to give the contractor a sufficient window of non-peak traffic flow to accomplish the work activities. Putting restrictions on the 20

contractor that require night work results in construction activities being accomplished in a more challenging period of the day. However, this does not necessarily result in less safe work zones compared with the alternatives. Several excellent resources are available to assist the STA in planning night construction activities, including NCHRP Report 475 and NCHRP Report 476 (19, 20). NCHRP Report 475 presents procedures to assist highway agencies in determining whether to perform nighttime construction or maintenance. NCHRP Report 476 contains guidelines for design and operation of nighttime traffic control. STAs in many locations are often balancing the tradeoff of completing a project or phases of work before winter conditions occur to prevent work during adverse weather conditions. Acceleration strategies that allow for a less restricted traffic flow condition through the winter certainly have positive safety benefits to the public. However, the reality is that construction work zones are inherently more dangerous than normal free flow traffic conditions. Narrow lanes, barriers, traffic merges, restricted ramp merges, fewer total lanes, distractions from the construction activities and additional signage are all elements experienced in a work zone that make the environment more difficult for the driver. This is especially important as one looks to the future with increasing traffic volumes on the systems that STAs are managing (21): • Between 1982 and 2002, vehicle miles traveled increased by 79%, while highway lane miles only increased 3%. • An estimated 3,110 work zones were present on the National Highway System (NHS) during the peak summer roadwork season of 2001. • Motorists encountered an active work zone 1 out of every 100 miles driven on the NHS, representing over 12 billion hours of vehicle exposure to work zones during 2001. • Highway workers spent 246.4 million hours working on the NHS during the year 2001. • In 2005, 1,074 fatalities resulted from motor vehicle crashes in work zones. This has grown from 1,028 in 2003 (a 4% increase) and 693 in 1997 (a 55% increase). • More than 41,000 people were injured in 2003 as a result of motor vehicle crashes in work zones. This has grown from 36,000 in 1996, an increase of 14%. • 53% of work zones are designated as day work, 22% as night work, and 18% are active all day or nearly all day (18 or more hours). It is clear that the use of acceleration strategies will increase in the future. The challenge for transportation professionals is to minimize or eliminate any additional complexity by the use of acceleration strategies on projects and, if possible, to reduce overall safety impacts to the public and construction workers. The following four sections contain questions posed by the NCHRP Project 10-58(01) panel and the research team’s responses. Is there data available that compare work zone accident rates for contracts that incorporate acceleration clauses and similar contracts that do not contain acceleration clauses? The research team was not able to identify enough data to directly answer this question. There may be several reasons for that. Foremost is analyzing all the variations and factors that could affect the results so that a useful analysis would re- sult. Most construction projects where acceleration strategies are used are on higher traffic volume, larger projects that are very complex in nature. Traffic diversion, improvements to the parallel network, special traffic enforcement, and the complexity that there may be several acceleration approaches incorporated into an overall project are a few of the issues that make the matrix of variables large and result in a very complicated analysis. No one the research team interviewed was aware of any attempt to conduct such on analysis. Several interviewees indicated that, in their judgment, it is nearly impossible to conduct such a comparison. Even a pre and post construction accident rate analysis, without considering acceleration clauses that may have been incorporated in a project, is very complex. However, the research team was able to identify one STA that has performed this type of analysis, the results of which are summarized as follows: The Ohio DOT (22) has conducted extensive analysis of its construction work zones. While the data are not linked with the use of particular contract acceleration provisions, the findings in general are excellent for identifying how to improve safety in work zones. The analysis of 2002 and 2003 work zone crash data and free flow comparables indi- cated a 60% to 70% increase in accident rates during con- struction compared with the post construction condition. Ohio DOT analyzed hundreds of work zone crashes looking for abnormally high concentrations of crashes and identi- fied the following areas of focus for improving safety during construction: 1. Geometrics—Off ramp capacity, inadequate ramp merges, insufficient paved shoulders, and 2. Speed reduction. Further analysis to determine whether the use of an acceler- ation clause on the contract had a positive or negative impact 21

on safety is so complex that none of the STAs contacted were aware of any conclusive studies that would provide insight on I/D impacts on safety. If data is not available or is incomplete, what judgment and experience do transportation professionals offer relative to the use of acceleration clauses as it relates to safety? The relationship between exposure time and work zone accidents was discussed with all the traffic professionals inter- viewed. The consensus was that, all other factors being equal, reducing construction time will reduce exposure time, which will correspond to fewer accidents overall. Acceleration strate- gies are normally used on higher traffic volume facilities, and all interviewees supported the concept of reducing the overall work zone exposure time. What strategies and considerations can a STA use to reduce risk to the motoring public, pedestrians, and construction workers on accelerated projects? During the interview process a number of helpful ideas were identified: • Apply good traffic control principles applicable to all con- struction work zones. Contracts containing acceleration provisions are no exception. Ensure that traffic control requirements do not create a situation where the contrac- tor can take short cuts on traffic control to gain speed of construction. This is largely a function of having good traf- fic control standards and requirements that the contractor must follow. • Be aware of potential secondary impacts on the parallel highway network around the project. Depending on travel time and traffic flow condition restrictions resulting from the project activities, route diversion may occur. On many projects, some diversion is desirable to maintain acceptable levels of service through the project. A network level analysis should be done so that appropriate countermeasures can be taken on diversion routes. • The actual safety impact becomes more complex on higher traffic routes. Carefully analyze the anticipated traffic volumes through the project work zone and match the available capacity to the acceleration boundary conditions. • Numerous technologies can be effective such as rumble strips, speed monitoring displays, facilitating merge oper- ations, smart work zone technology, drone radar, and improved nighttime traffic control practices. • The use of cameras to enforce speed, if permitted by state law, is a good alternative to uniformed officers. In many states, it is much more difficult to hire officers than it was 10 years ago. How do extended work hours impact construction worker safety? Research conducted in the field of occupational health reveals that extended work hours (either per day or per week) have a significant impact on the occurrence of work place accidents (23). A summary of the findings from a study con- ducted from 1987 to 2000 by Dembe et al. reports the following: • Working 60 hours per week or more increases the risk of injury by 23%. • Working 12 hours per day or more increases the risk of injury by 37%. It should be noted that this study was not specific to the construction industry. More than 110,000 job records were analyzed across all types of occupations. Because construction ranked in the top third of industry classifications for occur- rences of injuries, the actual increased risk due to extended hours is likely greater for construction than the overall com- bined rates reported. Although hard data are not available, there appears to be no safety related reason not to use appropriate strategies to accelerate construction contract completion. In fact, there is likely a positive public safety benefit to completing a project in the shortest possible time and having the improved facility in use sooner. Each project needs to be planned so that the traffic management requirements and the use of contract acceleration strategies are consistent. Project and network level reviews of anticipated traffic flows may be helpful in identifying potential secondary impacts. The contractor should be required to follow accepted standards for traffic control even when work is being accelerated. Acceleration strategies that rely strictly on extended work hours pose an increased risk of work place injury. Contractors and STAs should be encouraged to use management tech- niques that mitigate this risk. These techniques might include the following: • Multiple shifts • Replacement crews for alternating weekends • When project conditions permit, use of additional equip- ment and human resources to increase production instead of extended hours 22

Additional Comments Received from Contractors during Phase II Part of the Phase II research efforts involved soliciting feedback from contractors regarding their perspectives on time-related I/Ds. Additional comments that have not been included in the discussion of I/D impacts on project factors are provided in the following list: • “Road user fees, increased accident rates in construction zones and public perception are a value to STAs. Shorter construction durations generally are viewed positively by the General Public which may lead to increased funding for roads and bridges. This benefits both the STA and the Contractor.” • “In general I think A+B I/D projects are a good tool as long as parameters for things that are beyond the contractors control are looked at for both the incentive and disincentive. In the time frame that a contractor has to bid a project, it is impossible to look at all situations that may arise and the owner should bear some of the risk.” • “The STA should realize that I/D provisions are “insurance” to protect contractor markups. These cost reimbursements are for the inefficiencies and unproductive nature of compressed schedules. Incentives are not a bonus. STA should not under- estimate the impact of I/D provisions on their employees on the line. They work off peak hours without shift premium or compensation while the contractor’s employees may be com- pensated for the same time. They see the contractor receiving an incentive while their work is not recognized or rewarded.” • “I/D provisions bring more focus to management, supervision, and workers. Shortened durations for activities allow more focus for safety issues.” • “Any argument that a Contractor will lower their safety stan- dards for an economic return is a weak argument.” 23