Stable Maintenance Task Scheduling: A Bi-Objective Robust Optimization Model

What is it about?

This study aims to introduce an original concept entitled “stable maintenance tasks scheduling”. It is a useful contribution to the literature as it incorporates Robust Optimization (RO) concept into the maintenance and repair tasks scheduling problem. Developing a novel robust Bi-Objective Mixed-Integer Linear Programming (BOMILP) model to support the framework, where the maintenance tasks duration is uncertain, makes the contribution of the paper novel and unique. The main objective is to optimally schedule the maintenance and repair tasks in a minimum cost, while guaranteeing the stability of the scheduling. To provide more realistic scheduling, planning horizon is devided into some pre-purchased time buckets. The goal is to schedule the tasks in the available time buckets corresponding to a set of planned maintenance stop periods and preventive maintenance actions. To solve the problem, different Multi-Objective Decision-Making (MODM) techniques are applied to the proposed model. The best solution technique is then selected corresponding to such criteria as total cost, total float, and run time, by employing Displaced Ideal Solution (DIS) method considering three different-sized examples. It is revealed that Weighted Goal Programming (WGP) can generate high-quality solutions. Results further show the superiority of the proposed framework in maintenance tasks scheduling taking into account their nominal frequency. Besides, a negligible amount of objective functions variations, despite the uncertainty of the tasks duration, is indicative of achieving the main goal of the research in achieving the stability of the scheduling. However, task duration is revealed as the most sensitive parameter to the objective functions.

Featured Image

Why is it important?

An original concept entitled “stable maintenance tasks scheduling” is introduced. Dealing with the uncertainty, a novel robust bi-objective MILP model is formulated. This is the first to obtain expected maintenance cost and tasks total float simultaneously. The best exact solution technique is selected, using Displaced Ideal Solution (DIS) method. Different-sized examples are investigated to demonstrate the applicability of the model.