Optimization of number of checkouts in retail chain using imitational modeling with means of Python programming language

This paper explores the optimization of checkout systems in retail supermarkets using simulation modeling. The study employs queuing theory and simulation methods implemented in Python's SimPy package to evaluate and improve the efficiency of checkout operations under various scenarios. The model incorporates customer flow intensity, checkout service rates, and permissible queue wait times, which are dynamically adjusted based on the location type (transport hubs, central districts, or residential areas), day of the week, and peak or non-peak hours. Data sourced from Google Maps on hourly supermarket occupancy levels is used to simulate real-world scenarios. The analysis identifies optimal checkout numbers for different conditions by minimizing total costs, which include operating expenses for checkouts and losses from customers abandoning queues due to excessive waiting times. The results highlight significant differences in optimal checkout numbers depending on supermarket location and customer behavior patterns. For instance, transport hubs require fewer checkouts due to lower peak loads, while central districts demand higher numbers to manage consistent customer flow. Additionally, residential area supermarkets benefit from increased evening checkouts to accommodate after-work shopping surges. Sensitivity analysis reveals that the most critical factors influencing optimal checkout numbers are the maximum allowable wait time, service rate (dependent on average basket size), and customer flow intensity. These findings provide practical recommendations for retail management, suggesting adaptive checkout allocation strategies that align with customer flow patterns and service expectations. The proposed methodologies enhance decision-making and cost-efficiency, offering valuable tools for integrating queue optimization within supermarket management systems.