Development of Decision Support Platform for Designing Warehouse Operations

Logistics is one of the future industries that will be a driving force for Thailand’s economy. However, the logistics cost in Thailand is higher than in several countries in the world, including those in Southeast Asia. One activity that significantly affects the logistics cost is warehouse operations. Since 95% of the existing warehouses in Thailand are traditional ones; therefore, developing end-to-end smart warehouse solutions by utilizing technologies including digital, artificial intelligence, intelligent electronics, robots and automation will potentially reduce the logistics cost in Thailand. Although the automated smart warehouse can be designed by scenario modeling using ARENA, FLEXIM, or TECHNOMATIX, these simulation software programs are costly. Their technical specifications for a warehouse simulation are also limited. These drawbacks limit suggestions of amount of material handling equipment, number of operators, put-away policies and locations, picking policies, locations and paths. These suggestions are considered to generate system cycle time, capacity, and performances (i.e., lead time, waiting time, labor productivity, utilization of material handling equipment) needed by customers. This project adopts the idea of pharmacy automation, which constructs a simulation model using What-if scenarios on Microsoft Excel to suggest technology specifications that fit the hospital’s system performances. This project also adopts the idea of the Internet of Things (IoT)-based smart warehouse solution development that developed a simulation model using Greedy algorithm to find the shortest-path of wave picking in the test factory (TRL 4). Both ideas are integrated and applied to expand further in the development of the decision supporting platform for a design of smart warehouse operations. The first step of the platform operation is data mining, to screen and convert the by-transaction raw data into informative data, followed by modeling and optimization.  Multi-objective performances will follow target customers’ requirements and exhibit superior multi-dimensional characteristics over the current commercial software programs. Nevertheless, it is necessary to evaluate the platform with diverse target customers to ensure flexibility and stability before distribution, and to apply IoT-related technologies in developing the smart warehouse prototype to collect dynamic behaviors of products in a smart warehouse in real-time or near real-time to be used as the input data to the platform. The IoT data is significantly greater and more refined than one collected manually. This high-quality data thus helps perfect the platform validation process, ensuring the effectiveness of the platform utilization in actual warehouse operations proposed.