Warehouse With Bins

Challenge

This client wanted to optimize the time it takes for a set of operators to do the picking process based on a list of 15 items. The strategy to use was to locate the products with more demand in better positions and compare the resulting time with the time that is currently needed to do the same process.

Solution

The solution was developed in AnyLogic using exclusively the process modeling library. The location of each product was defined in an Excel database and the picking list (list of items to pick up in one run) was defined randomly based on the statistics of a few samples of real picking lists. Figure 1 shows the 2D animation.

Figure 1 – Warehouse 2D simulation

In these cases, the normal procedure when developing the simulation model is to generate a comparative experiment and visualize the results there, nevertheless, since the assumption was that a product was always available to pick, the simulation was created as a race between the group looking for products in the old bins and the group looking for products in the new bins. This allowed us to clearly visualize how and why the differences in time were so significative.

Outcome

The conclusion of the model was that the new positions saved around a minute of time in average for each picking when using 2 workers in the warehouse. Nevertheless, when reaching 4 workers, the new positions made no difference at all and were even worse when increasing to 5 workers. Optimizing the positions in a warehouse to minimize time is in fact a very complicated mathematical problem so it was very useful to visualize clearly in which cases the new positioning of the products was better.

Project Features

• Industry: Warehouse
• Model: Discrete Events
• Duration: 2 weeks