Moving to Risk-Based Highway Asset Maintenance Using Elicited Data

Abstract

This paper presents the journey that the UK’s largest highway asset management organization undertook to move to risk-based and data-driven maintenance decision-making, while maintaining the trust of maintenance managers and despite significant gaps in its asset data. It presents an application of the Sheffield Elicitation Framework approach for obtaining data from subject matter experts and approaches considered for aggregating the elicited data to maintain modeling quality. It outlines the approach taken to modify the elicited data with all available asset information through risk modifiers. It presents validation results for the aggregation method and describes how the outputs of residual risk given a maintenance scenario, and risk reduction per pound spent for maintenance activities, were arrived at. It then shows how they are being used as part of the decision support tool (DST) to optimize cyclic maintenance spend. The DST allows National Highways to maintain their level of service to customers whilst saving millions of pounds a year for UK taxpayers and achieving their asset management objectives, which is particularly pertinent when there is a significant focus on making the best use of limited resources. Because of the methodology it uses, it is also a knowledge management tool that represents a shift from fragmented decision-making based on individual maintenance managers’ tacit knowledge to utilizing the combined wisdom of all maintenance managers in addition to all asset data that is available. The paper concludes with a summary of opportunities for further work.