Elektrik dağıtım şebekelerinde etkin sayaç okuma planlaması için kümeleme tabanlı sezgisel yöntemler

In energy distribution systems, the measurement of customer consumption is an important task for utility companies. In Türkiye, electricity meters are periodically read as part of an operation that requires field workers to record the data manually. This meter reading planning operation constitutes a complex problem. This paper addresses the meter reading planning problem faced by an electricity distribution company responsible for several cities, including metropolitan areas. The problem is examined as a case study using real data. Due to national regulations, to ensure consistent billing, the company must read all customer meters on time. A large workforce is employed to ensure that all meters are read on a monthly basis. The objective is to minimize both the number of workers and the total distance traveled while completing the task within a limited time frame. Due to the NP-hard nature of the problem, a clustering based heuristic approach is developed to provide a sustainable and efficient plan. This solution uses clustering algorithms, local search algorithms, and mathematical optimization methods to generate efficient routes for workers. The performance and applicability of the algorithm are tested on real data from residential zones. The final output of the algorithm is intended to be integrated into future planning processes of the utility company, potentially leading to significant improvements and a reduction in the planning process time.