Phoenix Analytics is a Turkish consulting company and provider of simulation solutions for various industries. They were involved in developing a warehouse optimization project for one of the biggest ice-cream and frozen desserts manufacturers in Europe and Asia.
Phoenix Analytics engineers needed to create a simulation model of a cold store to help decision-makers for an ice cream manufacturer understand how to more effectively use resources.
The warehouse stores frozen products that require specific storage area temperature conditions. Temperatures in the storage area are affected by storage related activities, such as putaway and picking. For these areas of activity, three types of warehouse operators offered opportunities for optimization:
Engineers also had to follow several restrictions and rules while creating a model:
- First in First Out (FIFO) product storage
- All rack depths to accommodate the same product types
- Waiting time limits for pallets in the loading area
- Limits on the number of pallets in the loading area
- Limits on the number of pallets in the container area
Engineers used AnyLogic as warehouse simulation software. They developed a model that allowed flexible usage of various scenarios and the changing of cold store parameters depending on the season, rack load, staff shifts, etc.
Forklifts and collectors were classified as inbound and outbound resources. The priority for inbound and outbound dock usage could be easily changed to see the consequences and the impact of each setting.
In the model, the developers implemented a heat map color indication system for the warehouse racks. The heat map could be set up to present different information. Set to present for product type, the heat map is useful for analyzing the location of each product type for inbound and outbound operations. Setting the heat map to show the number of pickings makes it possible to monitor the racks with the most or least demand.
Considering numerous parameters, the warehouse optimization model determined the shortest travel distance for the forklifts and the collectors and showed where best to store pallets. It also showed the current state of fulfillment, the current state of resource utilization, and other statistics that help stakeholders in decision-making.
For a simulation run, engineers could use real data from different periods of operation to analyze performance.
The resulting statistics from a simulation experiment with the model include whether a scenario is successful or not for different combinations of resources as well as resource utilization. It is possible to see when service levels and KPI will not be met and to evaluate how different combinations of resources might perform.
The configuration that used the least resources while achieving the desired performance levels was chosen by the clients for their warehouse optimization needs.
The case study was presented by Ali Yoğuran, of Phoenix Analytics, at the AnyLogic Conference 2021.
The slides are available as a PDF >>