Purpose To build and design a simulation model to mimic customer behaviour using discrete event simulation. The model will incorporate a multi-channel queueing system. Thus, multi-servers will attend to customers in line. This improves service quality and avoids queues. Also, the simulation model will help identify gaps in the restaurant. Design/methodology/approach The research utilizes Arena simulation software to develop a model that simulates a typical South African restaurant. The model focuses on customer behaviour, operational flow and resource management, including staff and equipment. The simulation logic models customer flow through the restaurant. Upon arrival, customers are either seated immediately or wait in a queue if no seats are available. Once seated, customers place orders, which are passed to the kitchen for preparation. The time taken for each process (e.g. seating, ordering and food preparation) is recorded and analyzed for potential bottlenecks. Findings The use of discrete event simulation (DES) in conjunction with Arena simulation software offers a practical method for improving restaurant operations in South Africa. By modelling customer behaviour and operational processes, restaurant managers can identify critical bottlenecks and implement changes to improve efficiency, especially during peak periods. This study demonstrates that careful attention to staff scheduling, resource allocation and the layout of restaurant processes can greatly enhance customer satisfaction and operational effectiveness. The simulation results can serve as a decision-support tool for restaurant owners to test various strategies without disrupting actual operations. Research limitations/implications Further research could expand on this work by incorporating machine learning to predict customer behaviour trends or develop more detailed models that account for external factors such as load shedding, economic variability, or seasonal changes in customer patterns. Moreover, more optimization algorithms will be explored to identify an optimal solution to improve service quality to customers. Also, the Monte Carlo simulation will be included and embedded within a discrete event simulation. Furthermore, the future of DES in restaurant simulations lies in improving the realism of human behaviour models and integrating new technologies such as integration with artificial intelligence, real-time simulation and decision support and hybrid simulation models. Practical implications The application of DES in modelling human behaviour in restaurants offers significant advantages in operational management. As the restaurant industry continues to evolve, leveraging such simulation techniques will be crucial for maintaining competitiveness and improving customer experiences. The interaction of people at a restaurant, from when a customer arrives and places an order until they receive their food, was modelled using discrete event simulation. The simulation can be used to assess the restaurant’s performance, better comprehend the situation, and assess any new enhancements. The simulation’s overall effectiveness can be predicted using the model, along with the potential duration of an order and the impact of system changes. Social implications The baseline scenario revealed that customer wait times were acceptable during off-peak hours but significantly increased during peak times. In the peak hour scenario, customer abandonment rates (customers leaving without service due to long wait times) rose, suggesting that restaurants should implement dynamic staffing strategies to manage busy periods. The staff shortage scenario highlighted the critical role of adequate staffing in maintaining service quality, as both wait times and staff utilization were adversely impacted. Originality/value To the best of our knowledge, this is the first research on modelling human behaviour using discrete event simulation for South African restaurants.
Obagbuwa et al. (Tue,) studied this question.