Simulation-Based Order Management for the Animal Feed Industry

Animal feed production constitutes a significant market in today’s agricultural sector, with an annual turnover of 55 billion euros within the European Union in 2020. Nevertheless, feed logistics still suffer from low digitization and manually coordinated supply chains. These factors lead to high transportation and product costs for customers and retailers by inducing short-term orders that often disregard current price developments.

This article presents a simulation model for feed supply networks consisting of a number of customers, retailers, and manufacturers which are agents and placed on a Geographic Information System (GIS) map.

The simulation model proposed a fuzzy-based decision strategy for customers to decide when to order specific products. Moreover, it described a possible decision strategy for retailers to optimize their transport routes by selecting viable manufacturers. The agent-based design applied in the model allowed setting up different optimization strategies for each type of agent separately.

In this article, you will find a comparison of two strategies and see how the retailer’s travel distance and customers’ product costs differed. The evaluation showed that the proposed decision strategy could reduce costs for feed and, depending on the supply network structure, reduce delivery distances for feed retailers. The model also used the grasshopper JSPRIT external Java library imported to the AnyLogic model which allowed to solve several pickup and delivery vehicle routing problems.

The current implementation uses a simple linear regression to predict upcoming feed consumption from the last n data points. Future work will focus on integrating real-world data and developing viable demand, capacity, and price prediction algorithms. It was supposed to incorporate the selection and delivery of raw materials to manufacturers in the early morning and by evaluating additional decision strategies. The modular setup of the simulation allows, e.g., to try other heuristics or optimizations. Furthermore, the simulation model will be used as a training platform for reinforcement-learning policies, which could offer a powerful alternative to manually designed decision strategies.

The simulation model of the logistics system
The simulation model of the logistics system

Verwandte Posts