Multicriteria DSS with Historical Database for Attenuator Selection

Abstract

A computerized information system has been developed for the management of field data on vehicle impacts with crash attenuators. The system has been developed for an intranet platform with a Java-based interface. A central Oracle database contains a variety of data describing site conditions, maintenance history, and accident characteristics. Several standard reports can be generated, summarizing attenuator usage, life-cycle costs, and accident outcomes under a variety of site or crash conditions. In addition, a multicriteria decision support system uses historical data to rank attenuators based on their past performance under specific site conditions. This application supports the design and selection of attenuators for new locations; it can also be used to analyze current placement recommendations for common scenarios. Use of the system will allow highway safety personnel to better understand the performance and applicability of various types of crash attenuators. Further, the research extends the concept of a decision hierarchy by: (1) using deterministic and heuristic importance functions to vary the weights to reflect changes in site conditions; and (2) using a possibilistic matching algorithm to include historical data that partially matches the user-input site characteristics in the performance evaluations.