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.
6 This chapter examines literature related to estimating contract time methodologies. The focus of this section is on literature related to estimating contract time in general rather than specific state practices, which are covered in chapter four. TYPES OF CONTRACT TIME ESTIMATION PROCEDURES Taylor et al. (2013) examined published STA procedures for estimating contract time for highway con- struction projects. The tools examined were categorized into one of the following categories: (1) archived production rates (the tool relies solely on production rates for critical activities), (2) predetermined logic (the tool uses predetermined schedules and separate production rates), or (3) integrated schedul- ing (the tool has an integrated production rate and schedule logic based on bid item quantities). Table 1 shows a sample of archived production rates (output per day) for typical highway construction activities published from different STAs and FHWA (additional activities are shown in Taylor et al. 2013). The production rates were accumulated from state and federal publications. To prepare a contract time esti- mate, the estimator would create a bar chart using the critical activities and job logic that the estimator develops based on project size, characteristics, etc., and use the bar chart to estimate contract time. A system that uses predetermined job logic provides the estimator with a predetermined bar chart or criti- cal path precedence relationships based on project type. The relationships specify sequences between generic activities. The estimator would then use activities similar to those shown in Table 1 to prepare a project schedule. The estimator would then use bid item quantities and production rates to calculate activity durations and use the schedule to calculate the project duration. An integrated system uses software packages to develop a bar chart schedule based on bid item quantities, activity production rates, and generic precedence logic. The estimator inputs design quantities into a database system that calculates activity durations and produces a bar chart schedule. Figure 2 shows an example of an input screen for an integrated scheduling system and Figure 3 shows an example output schedule for an estimate project. Of the 50 STAs, 29 contract time determination procedures were made available for examination. The categorical break down was as follows: 48% use some form of integrated scheduling, 28% use archived production rates, and 17% develop a contract time based on predetermined logic. A system considered to be in the integrated scheduling category would use both archived production rates and predetermined logic. A procedure in the archived production rate category would not make use of a predetermined logic, which would eliminate said procedure from being in the integrated scheduling category. The additional 7% accounted for the two procedures that did not necessarily fit into one of the three categories. An examination of STA contract time procedures by Stoll et al. (2006) and Williams (2006) yielded similar conclusions. Some form of archived production rates were found in most of the procedures examined in the literature review; however, procedures included in this category were typically limited on any fur- ther method to aid in contract time determination. The second category used was for states that had implemented a procedure using predetermined scheduling logic. These procedures could have a pre- determined logic for work scheduling and phasing while using another method for determining activ- ity durations. Procedures examined in the review often had templates associated with project types commonly encountered in their state. Templates could range from a couple to more than a dozen chapter two LITERATURE REVIEW
7 options and each could have a different logic and/or production rate associated with the individual work activities. Integrated scheduling was the most abundant category utilized by STAs for determi- nation of contract time examined in the literature review. Procedures that were categorized here may have also used predetermined logic in combination with archived production rates; however, further action had to be seen that showed a way of integrating the multiple components that create a contract time. It appears that the trend is toward states having a method that involves integrated scheduling for determination of contract time. With each category, there are many possibilities for using each method; some procedures are developed within the department, whereas others may pursue more commercially available pro- cedures that are already structured to perform scheduling tasks. Combinations of the methods are seen in an attempt to create a customized procedure to best suit each departmentâs needs. This could range from a way to input current productivity rates, to determining logic based on certain aspects of a project, to inputting working day calendars based on holidays and weather conditions in a given region. Some programs found Microsoft software such as Access, Excel, and Project useful where variability in inputs is fairly easy, but complex data interaction can be limited; others use procedures developed by professional software developers such as Primavera and Field Manager, which can create much more complex components, but tend to limit user-defined inputs that may vary from project to project. CONTRACT TIME ESTIMATION ACCURACY Williams (2006) noted that although organizations such as the Association for Advancement in Cost Engineering have published classifications for levels of accuracy expectations of cost estimates as the design process progresses, the construction industry has not published time estimate expectations Activity Unit KY OK WA MN IN FL FHWA Clearing and Grubbing Acres 3 4 3 3 1.5 5 3 Roadway Excavation CY 5,000 2,825 1,500 2,500 1,600 Embankment in Place CY 4,000 2,825 1,700 2,200 3,800 1,097 Drainage Pipe LF 200 110 175 300 Box Culverts CY 30 50 50 10 Structure Excavation CY 300 2,825 80 Piling LF 300 300 300 250 Sub-grade Stabilization SY 8,000 2,500 4,000 Stone Base Ton 1,500 310 2,000 1,500 800 1,600 900 Asphalt Base, Leveling, and Wedging Ton 1,200 1,000 2,000 500 4,050 Curb and Gutter LF 500 300 Barrier Walls LF 500 1,045 200 Asphalt Surface Ton 1,000 900 1,000 1,000 900 Guardrail LF 1,500 1,000 400 400 750 Finish Seeding SY 4,000 11,616 48,400 2,500 23,500 12,100 Pavement Marking LF 10,000 10,000 15,000 6,000 36,960 37,000 TABLE 1 EXAMPLE OF ARCHIVED PRODUCTION RATES FOR ESTIMATING CONTRACT TIME
8 FIGURE 2 Example of an input screen for an integrated scheduling system. in as great detail. Stoll et al. (2006) examined the accuracy of an automated scheduling procedure in matching contract time estimates produced by STA personnel and found the estimates to vary across projects. Taylor et al. (2012, 2013) examined the accuracy of two published integrated scheduling tools for STAs and found that the average overall accuracy of the two procedures in estimating contract time was greater than +200%. The two main issues identified in the accuracy deficiencies of the proce- dures were the production rates and generic precedence logic. Taylor et al. (2013) found that published production rates in use at STAs were typically not specific to the STA (i.e., they were borrowed from studies at other states), were from older work samples that may not include modern construction meth- odologies, and did not take into account weather variations, project size, project characteristics, etc. In most instances the source of the production rate was not clearly identified. Generic scheduling logic was also found to be problematic because the logic typically did not reflect the activities that took place for a specific project or included default activity durations that were not based on production rates. These production rate and job logic issues could be magnified when the contract time tool was either
FIGURE 3 Example of a bar chart output of an integrated scheduling system.
10 FIGURE 4 Time to achieve cost certainty for alternative contracting methods (from FHWA Research Project Quantification of Cost, Benefits and Risk Associated with Alternate Contracting Methods and Accelerated Performance Specifications, DTFH61-13-R-00019 2013). used without appropriate judgment or completely ignored because of perceived inaccuracies (i.e., the tool is merely used to âcheck the boxâto be in compliance with FHWA procedures and the estimate is not used in setting contract time). ESTIMATING CONTRACT TIME FOR NON-TRADITIONAL PROJECT DELIVERY Although estimating contract time for non-traditional project delivery methods has not received a great deal of attention from researchers, non-traditional project delivery methods can impact the estimation of contract time as a result of the behavior changes they can elicit in project stakeholders versus a traditional design-bid-build delivery method. Design-build delivery methods can change the value placed on schedule performance resulting from the design-build team selection method. Gransberg and Senadheera (1999) examined design-build team selection methods for STAs and noted that different selection methods could result in different schedule incentives. Gransberg et al. (2008) found that 85% of respondents to a survey regarding the use of design-build identified the design-build delivery method as a possible strategy to reduce schedule durations. Best value procurement methods (e.g., cost + time bidding typically referred to as A+B bidding) provide com- petitive contractor incentives for both cost and time performance (Scott et al. 2006). This values schedule performance during the project procurement phase and would likely result in a different construction schedule than one that would be produced by a traditional design-bid-build delivery method. Construction management at risk pushes schedule risks to the contractor (Gransberg and Shane 2010). Alternative project delivery methods may improve project schedule performance. However, owing to the impact they have on project schedules, contract time estimation methods developed for traditional design-bid-build projects are likely to return unreliable results for projects using alternative delivery methods (design-build, CM/GC, etc.).
11 Another factor to consider in the estimate of contract time for non-traditional project delivery, or Alternative Contracting Methods (ACM), is the juncture when the design is sufficient to estimate the construction contract time beyond a low level of accuracy. According to FHWA research currently underway, cost certainty is achieved for ACMs at estimated stages as seen in Figure 4. This âcost certaintyâ should correlate to a project stage where contract time can also be estimated with higher accuracy.
