This paper aims to address the technician routing and scheduling problem (TRSP), a daily operational challenge faced by telecommunication service providers. The study is motivated by a real-world application in Saskatchewan, Canada, and aims to develop an effective and scalable model for technician assignment and routing under practical constraints. The design of the problem is unique because of the vast working areas in Saskatchewan.
A mixed-integer programming model is formulated to model the TRSP, capturing realistic constraints such as soft time windows, variable working hours, lunch breaks (LBs) and overnight shifts. Because of the NP-hard nature of the problem, the authors propose two metaheuristic algorithms – simulated annealing (SA) and genetic algorithm (GA) – to solve large-scale instances. Computational experiments are conducted using real data, and the metaheuristics’ performance is benchmarked against a commercial exact solver.
Results indicate that both SA and GA produce high-quality solutions within significantly reduced computation times compared to the exact solver. The GA consistently outperforms SA in terms of optimality gaps and solution robustness. The findings highlight the practical viability of using metaheuristics in large-scale technician scheduling problems.
The proposed approach offers telecom service providers a flexible and scalable solution for managing technician assignments efficiently while accommodating operational constraints. The metaheuristic algorithms can be integrated into decision-support systems to improve customer service and reduce scheduling inefficiencies.
This research makes two main contributions. From a modeling perspective, it incorporates various available technician working hours as well as LBs into the overnight TRSPTW model. From a solution methodology perspective, it develops two metaheuristic algorithms – an SA and a GA – to solve the overnight TRSPTW with the LB model. The effectiveness of these two metaheuristics is analyzed via computational experiments using real-world scenarios.
