Optimizing Warehouse Layout and Scheduling with the DOBT Algorithm

Traditionally, the integrity of orders has been overlooked in the parallel retrieval processes of multiple stackers. To address this, this paper proposes using an Order Tag to label all items within the same order, which is then used for scheduling retrieval tasks. The method for calculating these Order Tags influences the scheduling discipline of the ARS. To minimize average delay and ensure fairness, two new algorithms are introduced: the Dynamic Order-Based (DOB) Scheduling Algorithm and the Dynamic Order-Based with Threshold (DOBT) Scheduling Algorithm. To automate and expedite this retrieval process, a Smart Warehouse often utilizes an Automated Retrieval System (ARS) to manage and schedule retrieval tasks. Historically, the integrality of orders has been overlooked in the parallel retrieval processes of multiple stackers. To address this issue, this paper introduces the concept of an "Order Tag" to label items belonging to the same order. These Order Tags are used to schedule retrieval sequences for each stacker to optimize performance. Essentially, the design of the Order Tags dictates the scheduling discipline of the ARS. This paper proposes two new scheduling algorithms based on the concept of order integrality: the "Dynamic Order-Based" (DOB) algorithm and the "Dynamic Order-Based with Threshold" (DOBT) algorithm. Through simulations, it is demonstrated that DOB and DOBT significantly outperform existing methods such as First-Come-First-Serve (FCFS), Last-Come-First-Serve (LCFS), and Shortest-Job-First (SJF). Specifically, DOB and DOBT reduce the average order retrieval delay by at least 30% and decrease backlog pressure on downstream operations. Additionally, the DOBT algorithm allows for adjustment of a threshold value to control the maximum delay of orders, thus balancing fairness among all orders