This project presents a structured lab experiment designed to enhance students' learning experience by integrating Value Stream Mapping (VSM), simulation modeling, and digital twin technologies to analyze and improve the performance of manufacturing systems. This approach not only equips students with the technical skills to map, model, and test improvements in manufacturing settings but also deepens their understanding of contemporary methodologies in analysis and improvement of manufacturing systems. The experiment aligns with the objectives of the SQU-IEOM Simulation Competition, which emphasizes innovative educational techniques for improving manufacturing and supply chain performance.

The experiment is divided into four integrated phases to ensure a comprehensive learning experience and hands-on engagement with each of the methodologies. In the first phase, students will design and conduct physical simulation experiments to replicate a simple production line in a controlled lab environment. They will define a basic product and delineate the stages of its production, which will include multiple workstations. During the physical simulation, students will assume various roles some will perform the production tasks, while others will conduct time studies, document the process, and record data on processing and waiting times. This stage will help students capture and analyze real-time data, setting the foundation for subsequent simulation and digital twin development.

In the second phase, students will use the collected data to construct a current-state VSM, which visualizes the flow of materials and information across the production line. This VSM will highlight key performance metrics, identify sources of waste, and uncover inefficiencies within the simulated production system. The mapping exercise not only provides a baseline model of the current production state but also reveals areas where potential improvements can be made, reinforcing the lean manufacturing concept of waste elimination. This mapping activity will be supported by statistical data gathered from the physical simulation, giving students an analytical basis for identifying and quantifying inefficiencies.

The third phase involves developing a simulation model that serves as a digital twin of the manufacturing system, using the data and insights obtained from the VSM. In this step, students will employ SIMUL8 software, utilizing Stat::Fit tool to derive probability distributions for the workstation cycle times. This digital twin enables virtual testing of various improvement scenarios, providing students with a low-risk platform to predict the outcomes of adjustments before implementing them physically. This phase includes verification and validation activities to ensure the simulation model accurately reflects the real system’s behavior, achieved by comparing the simulation outputs with data from the physical experiment. Students will run a series of simulation experiments, testing potential improvement scenarios informed by the VSM analysis, such as adjustments to processing times or resource allocations.

In the fourth and final phase, students will conduct additional physical simulation experiments based on the most promising improvement scenarios derived from the digital twin model. By incorporating the changes proposed in the digital twin, such as adjustments to workstation layouts or process flows, students will validate these modifications in a real-world setting. Comparing results from this adjusted physical simulation with the simulation model outcomes will allow students to gauge the effectiveness of their improvements, bridging the gap between theoretical model predictions and practical implementation.

The implementation of this structured approach led to substantial improvements in the manufacturing system. Initially, a physical simulation of an assembly line for a simple product was conducted, using a base setup of four workstations. The analysis of the collected data from the simulation experiments revealed a bottleneck at Station 4, which had an average processing time of 16 seconds per unit, contributing to a total assembly time of 192 seconds per unit. ANOVA results further demonstrated that the differences in mean workstation cycle times were statistically significant. The VSM analysis indicated that 77% of this time is non-value-added, mainly due to waiting times. By developing a digital twin model and evaluating multiple improvement scenarios. The processing time at Station 4 is reduced to 10 seconds through line balancing, which increased the production rate from 150 to 257 units per 8-hour shift. The physical simulations validated these improvements, confirming an average cycle time reduction to 111.65 seconds, representing a 42% increase in throughput. These findings highlight the effectiveness of integrating VSM, simulation modeling, and digital twin to enhance manufacturing systems performance.

This lab experiment provides a multi-layered educational experience that immerses students in the iterative process of analyzing, modeling, testing, and refining manufacturing improvements. Through this approach, students gain valuable hands-on experience with VSM, simulation modeling, and digital twins skills that are increasingly relevant in modern manufacturing and supply chain management. Furthermore, this activity fosters critical thinking, collaboration, and problem-solving skills, as students work together to analyze data, develop models, and implement improvements.