Assessing repair and maintenance efficiency for water suppliers: a novel hybrid USBM-FIS framework

Abstract

A metropolitan area's water supply is imperative to the city's development. One of the main goals of the water supply utilities is to ensure the availability of water, as a lack of water would cause many social, political, or health problems. Therefore, water supply facilities must be in good condition, efficient preventive maintenance plans must be implemented, and the performance of the maintenance team monitored. In this paper, efficiency indices of Tehran water utility maintenance teams are investigated using different Data Envelopment Analysis (DEA) models. The final scores were then used as inputs to a Fuzzy Inference System (FIS) to assess the efficiency of these maintenance units. Two performance indicators based on DEA, "Availability efficiency" and "Repair time efficiency" are introduced for performance assessment. The Mean Time Between Failure (MTBF) and the Ready To Operate (RTO) are two desirable outputs that are considered in addition to one undesirable output: the Mean Time To Repair (MTTR). In addition, we suggest a new index named MRRW by combining the DEA efficiency with the RRW index. We introduce a novel approach based on DEA combined with FIS methods and the new factors for evaluating water supply maintenance systems, while most previous studies on measuring the efficiency of maintenance teams consider only limited aspects of performance measurement. Based on the results of our study, it became clear that the MRRW measures efficiency better than the traditional RRW measures. We present future improvement strategies based on the output of the FIS.