Risk Driven Analysis of Maintenance for a Large-scale Drainage System

Abstract

Gully pots or storm drains are located at the side of roads to provide drainage for surface water. We consider gully pot maintenance as a risk-driven maintenance problem. Our simulation considers the risk impact of gully pot failure and its failure behaviour. In this paper, we focus on two factors, the issue of parked cars and up-to-date gully pots status information, that may affect the scheduling of maintenance actions. The aim is to discover potential investment directions and management policies that will improve the efficiency of maintenance. We find that the "untimely system status information" is a dominant factor that weakens the current maintenance. Low-cost sensor technique could be a good development.

Multi-Objective Model and Variable Neighborhood Search Algorithms for the Joint Maintenance Scheduling and Workforce Routing Problem

“…Indeed, the workforce cost constitutes one of the highest costs in any organization. The major problems addressed in the first stream are technician routing and scheduling problem (TRSP) [8], technician and task scheduling problem (TTSP) [7], service technician routing and scheduling problem (STRSP) [6], and workforce scheduling and geographically distributed asset maintenance problems (GDMP) [2]. Cordeau et al [7] proposed a mixed-integer linear program and a solution approach for the ROADEF 2007 challenge.…”

“…A genetic algorithm is then proposed to solve large instances. Chen et al [2] studied the maintenance of gully pots or storm drains. They modeled a multi-period VRP, which considers the risk impact of gully pot failure and its failure behavior each day.…”

“…There is a relatively small amount of research combining preventive and corrective maintenance strategies since most of the papers examined dealt only with preventive scheduling, apart from [2,4]. Moreover, they did not consider the uncertainty of the breakdowns and did not deal with a multi-objective perspective.…”

“…For each of these six classes, we consider the horizon H = 100 h with a latest arrival time at the depot b n+1 = 200 h and H = 200 h with b n+1 = 400 h. The opening time is a 0 = 0. The maintenance parameters were generated as described in [4]: T pm i ∼ U c [5,10], Tcm i ∼ U c [15,30], [10,20], and σ ∼ t 0i * U c [2,5]. Only the parameter β ∼ U c [2,6] is generated differently compared to the one described in [4] so that it can be closer to reality.…”

“…The objective is to jointly consider the technical aspects of maintenance and the organizational aspects of operations management. To consider these two aspects, many works in the literature consider both maintenance and routing problems [2][3][4][5]. To handle this optimization problem, it is necessary to simultaneously investigate the maintenance scheduling and vehicle routing problem (VRP).…”