Data-driven capacity analysis of production systems: Insights from two case studies

Paper i proceeding, 2024

In this paper, we delve deeper into a method previously introduced by us for analyzing the effective capacity of production systems, a concept crucial to the efficiency of operations management. Departing from traditional methods that primarily rely on engineering specifications, physical factors, or time studies – all of which are prone to various limitations such as input sensitivity and behavioral biases – our approach stands out by leveraging the theoretical relationship among arrival rate, service rate, and waiting or throughput time. This innovative method enables the indirect estimation of service rate, a key parameter for understanding effective capacity, based on empirical data of arrival rate and waiting or throughput time, thereby circumventing the direct challenges and inaccuracies associated with traditional methods to capacity measurements.

To demonstrate the robustness and versatility of our methodology, we reference two previously conducted case studies. The first case study focuses on a healthcare setting, specifically an emergency department, analyzing patient flow and waiting times to determine effective capacity. The second case study transitions to a transportation framework, examining the impact of an automated gate services (ASGs) implementation at a seaport freight terminal by tracking the turnaround times of incoming trucks. These investigations not only showcase the method's applicability across diverse operational environments but also illuminate the inherent complexities and varied dynamics within logistics systems.

Our contribution to the field lies in advocating for a shift towards more theoretical, data-driven methods to capacity analysis. By comparing and contrasting the operational intricacies encountered in both healthcare and transportation settings, we underscore the practical relevance and adaptability of our methodology. This paper calls for a reevaluation of traditional capacity measurement methods, promoting a model that balances empirical data analysis with theoretical insights to enhance operational decision-making and efficiency across logistics systems.