Multidimensional Highway Construction Cost Indexes Using Dynamic Item Basket

Abstract

A highway construction cost index (HCCI) is an indicator of the purchasing power of a highway agency. Thus, it must reflect the actual construction market conditions. However, current methods used by most state departments of transportation are not robust enough to meet this primary goal due to (1) a significantly insufficient sample size of bid items used in HCCI calculation; and (2) inability to address the need to track highway construction market conditions in specific submarket segments such as, but not limited to, various project types, sizes, and locations. This study proposes an advanced methodology to overcome these apparent limitations using two new concepts: (1) dynamic item basket; and (2) multidimensional HCCIs. The dynamic item basket process identifies and utilizes an optimum amount of bid-item data to calculate HCCIs in order to minimize the potential error due to a small sample size, which leads to a better reflection of the current market conditions. Multidimensional HCCIs dissect the state highway construction market into distinctively smaller sectors of interest and thus, allow state Departments of Transportation to understand the market conditions with much higher granularity. A framework is developed to integrate these two concepts and a standalone prototype system, named the Dyna-Mu-HCCI System, is developed to automate the data-processing part of the framework. The historical bid data of the Montana Department of Transportation are used to evaluate the performance of the Dyna-Mu-HCCI System and measure the effects of the dynamic item basket (DIB) and multidimensional HCCIs. The results show an eightfold increase in terms of the number of bid items used in calculating HCCIs and at least a 20% increase in terms of the total cost of bid items used. In addition, the multidimensional HCCIs reveal different cost-change patterns from different highway sectors. For example, the bridge construction market historically shows a very different trend compared with the overall highway construction market. The new methodology is expected to aid state Departments of Transportation in making more-reliable decisions in preparing business plans and budgets with more accurate and detailed information about the construction market conditions. Further, the prototype Dyna-Mu-HCCI System is expected to significantly facilitate the HCCI calculation process and rapidly implement this new system.