Creating a Workforce Optimization Model for a Micro-Fulfillment Center

Creating a Workforce Optimization Model for a Micro-Fulfillment Center

Overview

Micro-fulfillment centers are small, sometimes highly automated facilities that can fulfill online orders for delivery as well as click-and-collect in-store pickups. They may be located at the back of an existing retail store or warehouse.

Orders can be picked using an automated storage and retrieval system or a combination of this and manual picking. This automated picking technology can improve the picking process, reduce labor costs, guarantee that orders meet the highest quality standards, and minimize the friction of shopping with in-store customers.

However, it is very challenging to organize the labor necessary for a system that includes both automated and manual aspects. Planning proper staffing is essential because, if done incorrectly, it can lead to the retailer’s inability to fulfill their customers’ needs.

Problem

Retailers needed to know the staffing required to operate their micro-fulfillment centers, and they also needed to understand how different scenarios would impact their ability to fulfill orders. For instance, it was critical for the retailer to know the impact of an automated picking station or decanting station going down, or if there was a bottleneck in any step in the processing of an online order.

Balancing the work of the machines and not understaffing or overstaffing is key, especially as staff need to be moved from picking to delivering at different high concentration points during the day.

Dematic uses automation, AI, and real-time data to help businesses improve their operations. They design and manufacture automated storage and retrieval systems, conveyors, sortation systems, and order picking systems. In addition, they help design, build, and support intelligent automated solutions.

In this project, they focused on setting up the operational readiness framework for grocery retailers with a micro-fulfillment center in the back of their store. Dematic worked with a number of different grocery retailers to better understand the necessary workforce optimization model. So, they were not solving the problem for one particular store but for many potential stores.

Solution

Dematic decided to use AnyLogic to create their workforce optimization model because it has low coding and the ability to combine the Material Handling and Pedestrian Libraries, which simplifies the simulation of warehouse operations and pedestrian flows within facilities. That's why the company has decided to use them together. Tutorials, available literature, and an online support community helped them get a grasp on the software and find answers to any questions they had.

In addition, they could create beautiful-looking simulations to show their customers and back up these visuals with real data for analysis and validation. Finally, they were able to upload their model to the AnyLogic Cloud, meaning that they could share the model with other teams within their company, who could then also experiment with it.

Several elements needed to be considered in the model, including inventory management, picking, order marshaling, and delivery. The process flow is illustrated below and shows the high level of complexity that the developers faced when designing the simulation.

Flow chart illustrating a micro-fulfillment process

Micro-fulfillment process flow (click to enlarge)

Key features of this model included the ability to upload customer order data and staffing schedules from an Excel file, the ability to adjust rates (picking, decanting, etc.), the prioritization of tasks by zone, and capacity constraints by zone. Importantly, customer data would include order data in real time.

After uploading the necessary files, users can adjust a number of different parameters in the model. These include the distribution of the orders that are going to be picked by automated systems as opposed to manual ones, the number of people working, the number of picking and decanting stations, and various rates. A probability distribution is being used for the various rates, so the simulation doesn’t just use the fixed numbers they put in.

The developers or retailers can also adjust the priority levels, which means they can identify some of the most important tasks. Then they can make sure that those are prioritized so that they don’t become a bottleneck. Finally, users can also adjust the length of time it takes to run the simulation.

There are four tabs on the top of the model that help retailers understand how it works:


Workforce Optimization Model for a Micro-Fulfillment Center

Results

The results are the data output, as illustrated below, that can be presented to potential retailer customers. It gives an hour-by-hour breakdown of how many people are working, where they are working, the total number required, the staff utilization, how many orders can be fulfilled with the number of people simulated, and other details.

One of the main strengths of the model is that it provides a visual, digital story that helps all levels of an organization understand the benefits of it. This workforce optimization model is currently a work in progress and may be developed further in the future.

The case study was presented by Ricardo A. Ugas and Stephen E. Hoffman, of Dematic, at the AnyLogic Conference 2022.

The slides are available as a PDF.



Похожие проекты

Другие истории успеха

Сборник историй успеха от пользователей AnyLogic

Скачать