Analytical Dashboard Designs for Evaluating and Monitoring Automated Pick System Performance in Fulfillment Centers

The supply chain has transformed largely over the past decade in the form of e-commerce, and a vast number of companies are investing in robotic automated warehousing solutions to increase their speed, accuracy, and precision. Decisions to invest millions of dollars into automation must be made thoroughly, with great caution, and must be practically backed up to serve the needs of the company and the customers. Major retailers like Walmart, Amazon, and Target have all highly invested in robotic solutions for their fulfillment centers. This paper focuses on fulfillment centers as these large facilities are the new backbone of the complex supply chain. The concept of fulfillment centers caters directly to the retailer’s physical stores (if existent) and to the customer, who is the single most important link in the entire supply chain for the retailer. The pick process in a fulfillment center is one of the most important operations that drives customer order fulfillment. Traditional pick processes involve manual walks performed by workers that pick products. To replace this painstakingly slow and laborious process, retailers have partnered up with companies specializing in robotic solutions to implement systems such as part-to-picker, part-to-robot, and many more that drive speed and automation in the facility. This paper aims to introduce data analytics and data visualization that aids retailers to monitor pick systems at fulfillment centers. Post installing these complex and expensive systems in place, there exists a need to continuously monitor and evaluate the performance of these systems. The use of dashboards is the most apropos method of showcasing important metrics about these systems that are of interest to the specialized teams within the facility, as well as to top management. The motivation behind presenting analytical dashboards through this paper is to engage in the process of standardization. In line with the complexity of integrating dashboards to cloud servers for constant data pull, the designs of these dashboards should be clutter free, visually appealing, easy to understand, portray the right performance metrics, and flexible enough to adjust according to the operational need. Such dashboards are designed and presented toward the end of the paper with important metrics succinctly displayed on the dashboards.

Advisor Details: Dr. Paul Griffin Professor Harold and Inge Marcus Department of Industrial and Manufacturing Engineering