Estimating Highway Preconstruction Services Costs - Volume 2: Research Report (2016)

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.

59 C H A P T E R 6 6.1 Introduction This chapter focuses on the results of the research and the recommendations for future research that were recognized during the project. There were three separate research instru- ments used to reach conclusions: the literature review, the national survey, and the case studies from structured inter- views with agency personnel. 6.2 Conclusions Throughout the course of this research project, the research team was able to reach several conclusions. These are discussed in the following, in no particular order. There is a need for improved PCS cost estimating. The liter- ature has shown that underfunding preconstruction activities results in shifting problems that have not been identified or addressed to the construction phase, where the cost to correct them is much higher. • The implications of bad/no PCS cost estimating are cur- rently poorly understood: A number of the case study inter- viewees expressed the opinion that having a more accurate estimate of PCS costs would not add any value to the pro- cess. These opinions were based on the disparity between the cost of PCS and construction; roughly 3% to 6% for PCS and the remaining 94% to 97% for construction. Addition- ally, most indicated that there was little if any accountability for keeping actual costs within the amount budgeted for the preconstruction phase. This may account for the low level of confidence that respondents attributed to the quality of the existing data (e.g., there is no perceived reason to accurately post actual hours charged to a particular project because there is no monitoring of PCS costs). These observations reinforce the importance of crafting a PCS cost-estimating guidebook and educating practitioners on the importance of PCS cost estimating during the project development phase. • Terminology and data coding systems: No standardized terminology was found across DOTs for the various com- ponents of the project development process. Hence, gen- eralized estimating models and processes needed to be developed in the guidebook to allow DOTs to tailor them to their specific needs. This lack of standardization has also been seen in the management of preconstruction activities and data-collection methods within DOTs. Useful histori- cal information to analyze and estimate PCS costs is being recorded in multiple databases using different terminologies and coding systems, reducing the agencies’ ability to com- municate and share information among functional areas. This issue adversely affects the accuracy with which an esti- mator can assign various expenses found in the financial record to specific tasks in the PCS process. • Data quality: In many of the case study projects, the DOT personnel expressed doubt regarding the accuracy of the available data. Some of the DOTs have a sophisticated pro- cess for collecting the PCS data, but in all cases, the data depends on the diligence of the individual employees to accurately reflect the distribution of the hours charged to a given project in a normal day. These data quality issues have proven to be a critical factor affecting the develop- ment of effective PCS cost estimates. Thus, the PCS cost- estimating guidebook also provides some suggestions for DOTs to improve their data-collection techniques, making PCS information more accessible and easier to understand by all involved in the project development process. There is no elegant statistical methodology to remedy this issue. However, the result of this finding led the team to decide to use three different methodologies for modeling PCS costs (artificial neural networks, multiple regression analysis, and decision trees), which will yield three individual outcomes for any given data set. This permits the analyst to “bound the outcome” as there will be a low value, a high value, and a value in between, producing what is called a credible range in the risk-based estimating literature (Anderson et al. 2007). Conclusions and Recommendations for Future Research

60 Thus, the result is a worst, best, and most likely case for the project’s PCS cost. • PCS cost estimating is required when negotiating exter- nal consultant contracts. Federal law stipulates a “detailed cost estimate . . . with an appropriate breakdown of specific types of labor required, work hours, and an estimate of the consultant’s fixed fee . . . [is required] for use dur- ing negotiations” (U.S. Government Printing Office 2015). As a result, a PCS budget determined from a percentage of construction costs is not sufficient to use for negotiat- ing PCS contracts with external consultants. Top-down estimating approaches cannot be used in this application. Functional-level estimating is necessary for consultant negotiations. 6.3 Recommendations for Future Research Throughout the course of the research, the research team noted some areas that could be explored to further improve PCS cost-estimating practices: • The research identified the lack of similar terminology between agencies. Investigating the implementation of a standardized process to collect and use PCS cost data would enable information to be shared easily between DOTs. Les- sons learned could then be universally applied to improve all agencies’ practices. Another branch of this concept would be to develop a standard coding system for project factors.