The world’s leading automotive manufacturers use simulation modeling to optimize production line processes, improve scheduling, create forecasts, and integrate new technologies. Using simulation modeling helps them stay at the forefront of innovation and avoid costly mistakes.
Here are five case studies that highlight the different areas automotive manufacturers apply AnyLogic’s multi-method simulation modeling.
Improving automotive production lines
Engineers working on process optimization for Mercedes Benz’s Sprinter van production line gave a presentation at the AnyLogic Conference 2021. The presentation showed how they developed a simulation model that provided immediate improvements as well as ongoing optimization and decision support.
After replicating the production line as a simulation, the company’s planners realized a 5% efficiency gain by streamlining current processes. They also identified a further possible 5% efficiency gain that could be made by increasing automation.
In addition to the efficiency gains, the tool provided the planners with 90% of the calculations they previously had to do manually.
The vehicle factory simulation will now support the roll-out of automated guided vehicles that serve the production line and form the basis of a fully automated shopfloor logistics system.
On an Iveco Daily commercial van assembly line in Italy, global capital goods company CNHi implemented an industry 4.0, smart factory, approach to maintenance. The cost of a single minute of downtime could be more than $160k and improving maintenance of assembly line machinery would deliver savings.
To test the smart factory concept, engineers created a digital twin of the van chassis welding line. The digital twin coupled data from the real-world production line machines with their simulation counter parts. The result was a tool for production line analysis and for connecting with reinforcement learning that can predict assembly machine failures.
CNHi management used the project to test what-if scenarios and develop optimal maintenance policies based on data it provided.
Optimizing production line autonomous mobile robots
Automated guided vehicles (AGV) are used throughout manufacturing for transporting parts around shopfloors. Now, their more complex cousins, automated robotic vehicles (AMR), are becoming increasingly common. Thanks to the total cost of ownership often being lower for AMR than AGV, adopting these more capable automated vehicles is an attractive option. But, to be truly effective and efficient, they must have good policies governing their operation.
Tesla Material Flow Engineer and former BMW Group AMR researcher, Maximilian Selmair, develops algorithms for AMR task allocation. A case study of his work describes standard industry practice when deploying large-scale transporter fleets and demonstrates how AnyLogic cloud-based simulation helps develop optimal task allocation algorithms. A method developed from the Jonker-Volgenant-Castanon (JVC) assignment algorithm reduced the number of AMR required at a site by 30%.
Adapting to Cars as a Service (CaaS)
Simulation is not only used to optimize production line processes. A multinational US-based automaker understood that changes in the automotive industry could develop into an existential risk. The shift from car ownership to vehicles and cars as a service (CaaS) caused the automotive maker to work on becoming more innovative and resilient by changing its business processes.
The company approached the problem from a social theory point of view. Engineers mapped employee networks and features into a simulation model of business processes. AnyLogic’s multimethod capabilities allowed system dynamics and agent-based approaches, in combination, to accurately capture relationships between entities, their dynamic nature, and the adaptive-cycle theory. The open API of AnyLogic connected with Python libraries to facilitate in-simulation network analysis.
The result of the automakers project was a tool that increased business resilience and innovation by enabling more effective team management.
Predicting market demand in the United States
In a further example of simulation applied in the automotive industry beyond the realm of production processes, here is an example of market modeling for a world-renowned US motorcycle manufacturer.
The request for a 5-year strategic forecast of performance for marketing led to a simulation model of the entire US market.
The model captured market geography and demographics, as well as employment statistics and customer attitudes towards products. After careful validation against historical data, the model was used to align the company’s marketing and production policies.
From production line processes, through AI integrations and robot control algorithms, to analysis of human behavior, automotive manufacturers are using AnyLogic simulation modeling to develop their businesses.