A prescriptive analytics approach for evaluating two production systems: Simulation optimization algorithm

Production systems influence cost performance and carbon emissions. Environmental concerns compel companies to optimize energy efficiency in their production processes. This study explores the dilemma associated with fixed-time and fixed-lot systems during random disruptions and how these systems can improve performance. We employ simulation optimization models in business analytics using the discrete event simulation provided by the SimPy library within a Python environment. The study is based on the statistical analysis of data collected from 624,000 simulated hours. Our analysis reveals that a higher service level tilts the balance, favoring adopting a fixed-time production system in scenarios characterized by significant disruptions. A system with higher demand variability and lower standalone workstation availability (indicative of more variable production) tends to favor a fixed-time batch production approach. When workstations operate at low-capacity utilization combined with high standalone availability, the fixed-lot batch production system becomes more cost-effective. Overall, the fixed-time system demonstrates a superior capacity to accommodate higher production variability levels than the fixed-lot system. This paper contributes to the existing literature by providing simulation-optimization evidence to assess the relative efficiencies of fixed-size and fixed-time lot batch production systems. This paper considers the impact of random disruptions on operational efficiency within fixed-size lot batch production systems, highlighting the consequences of variability in lot completion times. The study also contributes to strategically selecting production systems to optimize energy usage in manufacturing processes.