SWUTC Research Project Description
Decision-Support Framework for Quantifying the Most Economical Incentive/Disincentive Dollar Amounts for Critical Highway Pavement Rehabilitation Projects
University: Texas A&M University
Principal Investigator:
Kunhee Choi
Texas Transportation Institute
(979) 458-4458
Project Monitor:
Dr. Young Hoon Kwak
School of Business
George Washington University
Washington, D.C. 20052
Funding Source: USDOT and State of Texas General Revenue Funds
Total Project Cost: $56,625
Project Number: 600451-00016
Date Started: 4/1/12
Estimated Completion Date: 3/31/13
Project Summary
Project Abstract:
In an effort to motivate contractors to complete construction projects early on high-impact highway pavement construction projects, state transportation agencies (STAs) including TxDOT have often used incentive/disincentive (I/D) contracts. However, determining I/D rates is extremely difficult due largely to the lack of systematic methods for helping STAs determine effective I/D rates. The primary goal of this project is to develop a novel framework for determining the most realistic and economical I/D dollar amounts for high-impact highway improvement projects. To achieve its goal, this project proposes an integration analysis including project schedule and the lower and upper bounds of the I/D contract. The lower bound is the contractor’s additional cost of acceleration, and the upper is the total savings to road users and to the agency. This project will provide the research community with the first view and systematic estimation method that STAs can use to determine the most economical and realistic I/D dollar amount for a given project–an optimal value that allows the agency to stay within budget while effectively motivating contractors to complete projects ahead of schedule. It will also significantly reduce the agency’s expenses in the time and effort required for determining I/D dollar amounts.
Project Objectives:
The primary objective of this project is to develop a novel decision-support framework that determines the most realistic and economical I/D dollar amounts for time-critical highway improvement projects.
Task Descriptions:
Task 1. Gather preliminary information and identify gaps in knowledge.
This task will provide the research team with a preliminary assessment of current state-of-the-practice with I/D provisions.
Task 2. Develop comprehensive time-cost tradeoff data from stochastic simulations sourced from real-world highway pavement rehabilitation projects.
This task will provide comprehensive time-cost tradeoff data for use in developing a stochastic model that estimates contractor’s additional cost of acceleration (lower bound of I/D). In this task, schedule simulations will be carried out using an innovative software tool called Construction Analysis for Pavement Rehabilitation Strategies (CA4PRS) based on contractors’ actual construction plans sourced from real-world construction projects. These simulations will generate schedule compression trend data that model the relationship of “time-cost tradeoff” on four different levels of resource use; that is, ordinary, 5 percent increase, 15 percent increase, and 25 percent increase in the number of resources per hour per team. Changes in cost in response to schedule compression will then be calculated based on a cost manual published and updated annually by TxDOT (or RSMeans unit price information). A set of contractors’ time-cost tradeoff data will be created on the four different resource usage levels by calculating changes in cost in response to schedule compression.
Task 3. Develop Stochastic Models for the lower bound estimate of I/D.
In this task, the relationship between time and cost will be plotted from the time-cost tradeoff data to determine an appropriate initial regression equation and a regression analysis will then be carried out to create a prediction model that determines the contractor’s additional cost of acceleration (lower bound of I/D dollar amount).
Task 4. Prepare Interim Progress Report.
An interim report will be prepared at the conclusion of task 3 for project monitor’s review and comments. Following review of the interim report by the designated project monitor, PI will make changes accordingly to reflect his comments on the project.
Task 5. Develop a stochastic model for the upper bound estimate of I/D.
This task will produce a stochastic model that estimates the total savings achieved for road users and the agency by accounting for heterogeneity of drivers’ value of time. The proposed new methods incorporate driver’s value of time as heterogeneous along with project schedule, contractors’ additional costs, and uncertainties in other key parameters. This task will also incorporate a Bayesian approach assuming unknown parameters to have distributions into jointly modeling I/D as a function of road user cost, contractor’s additional cost of acceleration, and other uncertain variables.
Task 6. Validate research results.
In this task, the robustness of the proposed framework’s reliability in estimating optimal I/D rates will be validated through several case studies with TxDOT’s actual incentive projects by looking at prediction errors.
Task 7. Prepare final report.
Any changes as a result of the validation process will be included in the final Guidebook.
Implementation of Research Outcomes:
One innovative way of reducing construction duration is to reward contractors with an early completion incentive bonus and levy fines for delays. Although use of Incentive/Disincentive (I/D) is increasingly common, State Transportation Agencies (STAs) often struggle to select the most appropriate I/D rates due largely to the lack of the proper analytical methods. There is an immediate need to develop a holistic framework that is more general and applicable to a variety of transportation projects for the determination of optimal I/D rates.
This project produced a new seven-stage decision-support framework that determines optimal incentive/disincentive (I/D) dollar amounts for time critical highway pavement rehabilitation projects—an optimal value that allows the agency to stay within budget while at the same time effectively motivating contractors to use their ingenuity to complete the projects earlier.
Products produced by this research:
Anticipated presentations: To TRB 2015, International Symposium on Automation and Robotics in Construction (ISARC) 2014, CRC 2015, International Conference on Construction Engineering and Project Management (ICCEPM) 2014
Academic Course Material: Research results incorporated into the curriculum for the graduate course Advanced Infrastructure Planning and Management in the Construction Science Department at Texas A&M University.
Journal articles in preparation: To ASCE Journal of Computing in Civil Engineering, Elsevier Journal of Automation in Construction, and ASCE Journal of Construction Engineering and Management.
Impacts/Benefits of Implementation:
This research is the first of its kind and expected to be a significant leap forward over current ad-hoc approaches that rely heavily on agency engineers’ intuition, historical data, and judgment, made primarily on the impact of I/Ds on road users. The proposed work provides research communities and industry practitioners with the first holistic view that they can use to determine the most economical and realistic I/D dollar amounts for a given project. Use of the proposed framework can help agency engineers and decision makers make better-informed decisions and allocate more realistic incentives when they consider the implementation of an I/D provision, which will result in more favorable cost-benefit ratios and better use of public funds. If the agency allocates an incentive smaller than the contractor’s added cost, this may keep competitive contractors from submitting a bid. Use of the framework can also benefit contractors bidding on projects that include I/D provisions because it can provide them with advanced knowledge of the balanced time-cost tradeoff amount required for acceleration. Critically, it will significantly reduce the agency’s expenses in the time and effort required for determining I/D rates.
Web Links:
Final Technical Report