This chapter analyzes and proposes methods for utilizing artificial intelligence (AI) in the development of smart warehouses in response to the crucial role and rapid expansion of smart warehouses. This study intends to produce value strategies and AI-powered solutions for the smart warehouse to increase warehouse process efficiency and bolster the logistics sector's competitive edge, particularly in Vietnam. In addition, this chapter presents a scale model of a smart warehouse system that uses cutting-edge technologies to manage a warehouse with the main objectives of improving operational efficiency, better meeting customer expectations, and eventually giving businesses in the logistics market a competitive advantage. The Eastern International University (EIU) smart warehouse testbed is an application testing facility that makes use of a 1:10 scale replica of a real system with features including automatic guided vehicles, RFID, a PLC system, and a warehouse management system. In one dual-rack system, up to 196 packets of three different sizes (0.5 kg [green], 1 kg [yellow], and 1.5 kg [red]) may be kept. A wide range of stakeholders, including students, professors, researchers, and businesses, benefit from the initiative. From an academic perspective, students benefit from a highly practical learning environment when dealing with the topics relevant to Industry 4.0 (I4.0) and digital transformation. The faculty, on the other hand, has a great environment for an in-depth study of supply chain management with a particular focus on warehouse management.

This study conducted a hardware design approach of critical smart warehouse key components with the desire to best support digital transformation and AI implementation. Specifically, three important constituents related to the three main areas of a smart warehouse (i.e., mechanical, controller, and data collection algorithm), including telescopic fork mechanisms for AS/RS stacker cranes, multilayer industrial controller, and an effective algorithmic data collection solution was created and demonstrated. The strict requirements of pallet storage smart warehouse comprise (1) mechanical mechanism: high payload, high positioning accuracy, low taping deformation, high 84acceleration transmission, and low noise; (2) controller: industrial standard, cloud database communication, and storage, accurate inventory, stock keeping unit planning, inventory management, circulating conveyor system, and control of incoming and exiting items; (3) data collection: high capacity with 1000 pallets storage, pallet circulation, and management code reassignment, and continuous operation; and the desire for the compatibility and possibility of complementing IoT and AI add-ins in the future has been considered and guaranteed. The results show that the input design parameters and constraints are well captured and assured.

This study proposed and implemented telescopic fork transmission mechanisms for automated storage and retrieval system (AS/RS) stacker cranes in the smart warehouse model. The strict initial requirements of the system comprise high payload, high positioning accuracy, low taping deformation, high acceleration transmission, and low noise. To realize these concerns, two versions of telescopic fork transmission mechanisms were approached and conducted. Specifically, the first version applies a cable roller combined with a timing belt transmission, and the improved version utilizes a timing belt. The experimental results demonstrate the superior performance of the enhanced version. The achieved results for the improved version include: the maximum taping deformation is 0.6 mm; the positioning accuracy is ±0.2mm; the backlash of the mechanism is 0.037 mm; the noise level is 43.6 dB for closed-door room operation. These performance parameters all satisfy the initial requirements. Moreover, the performance indicators of the improved version are mostly superior except for the noise level compared with the first version.

In this study, a multi-disciplinary cooperative project was implemented with the desire to close the gap between university education and industry, equip students with “work-ready” competencies, and the ability to work in the global and emerging technologies environment. Consequently, a project entitled smart warehouse with the participation of students from three-related schools was proposed and implemented. Through the project, the students developed a variety of competencies, including literature review, brainstorming, concept sketching and design, teamwork, peer/cross-evaluation, hand-on skills, interdisciplinary knowledge, problem-solving, project planning, modular design concept, design of experiment, etc. Specifically, through the survey, the average improvement in the Level of Confidence is 1.35 out of 4, and that of the Level of Knowledge or Skill is 1.29. Generally, the overall average improvement levitates from 2.20 to 3.51.

This paper presents the design and implementation of a miniature model of a smart warehouse system. It utilizes advanced technologies and digital technology to manage a warehouse with the primary goal of increasing operational efficiency, better serving customers’ needs, and ultimately creating a competitive advantage for enterprises in the logistics market. The EIU Smart warehouse system is an application testing center using a 1:10 miniature model (prototype) of a physical model with functions acting as a real system with advanced technologies such as Automatic guided vehicle, RFID, PLC system and warehouse management system. There are up to 196 packages with three different sizes (0.5 kg (green), 1 kg (yellow) and 1.5 kg (red)) that can be stored in the one dual-rack systems. The project serves various stakeholders, including students, faculty, researchers, and enterprises.

Responding to the vital role and robust development of smart warehouses, in this study, solutions for applying Artificial Intelligence (AI) to smart warehouse development, especially in Vietnam, were analyzed and proposed. The paper has investigated the factors affecting the application of AI for smart warehouse development in Vietnam through SWOT. Solution groups include (1) Investment decision support solutions that limit investment risks through solutions to test innovative approaches and algorithms, (2) Solutions for AI application in smart warehouse development, (3) Solutions to develop AI resources for smart warehouses. Therefore, by using SWOT analysis, this paper aims to generate value strategies AI-based solutions for the smart warehouse to improve the warehouse process efficiencies and strengthen the competitive advantage of the logistics industry in Vietnam.

Nowadays, warehouse optimization is one of the core components of logistics. With the development of artificial intelligence (AI) technology and the advancement of automation technology, building a smart warehouse is an important task. This paper presents the machine learning techniques and technologies for developing an intelligent warehouse. A reinforcement learning method is proposed to train a basic warehouse environment for an efficient storage policy. The experimental results help comprehend the building of the model of a neural network of the reinforcement learning algorithm and the characteristics of this technique. This study further helps to understand the basic concepts of machine learning techniques to develop an algorithm for smart warehouses.

In this paper, a smart warehouse testbed was designed and simulated. The system serves two purposes. It simulates a real-world industrial setting and emphasizes the significance of experimental research to assess the feasibility of new ideas in real-world situations. The suggested simulation model has five rack systems with a total capacity of 1000 pallets. An AGV moves in the middle of each rack system, picking up and returning pallets to both sides in the direction of movement. A circulating conveyor system is used in the proposed system to enhance transaction count data to enable intelligent algorithms such as machine learning. Accordingly, each pallet (attached with one RFID tag) is managed by a management code when stored in the warehouse. After taking out, that pallet can be cleared of the old management code, assigned a new code, and returned to the input conveyor of the system. One of the biggest challenges of the proposed system is ensuring the continuous operation at the inlets of the AS/RS AGV’s pallet receiving conveyors. The feasibility was demonstrated by conducting the simulation to build a polynomial regression function from which the value of Tseparator can be attained according to the given set of values (Vconveyor and Tstoring). In addition, when the set of values (Vconveyor, Tstoring, and Tseparator) is known, the time required to store the entire inventory may also be calculated.

Regarding a smart warehouse, building an industrial control system (ICS) that works effectively with the requirements is essential. An ICS, on the other hand, is a broad category of command and control networks and systems that support a wide range of industrial processes. They include SCADA systems, distributed control systems (DCS), process control systems (PCS), safety control systems (SIS), and other, generally smaller control system designs like programmable logic controllers (PLCs). In this paper, a typical 4 layers ICS, including a warehouse management system (WMS) software, was introduced to control all processes in the supply chain of a smart warehouse testbed. The WMS software initially achieved the set goals in terms of basic tasks. Furthermore, ICS worked well without incidents such as blocked or delayed flow of information, inaccurate information sent to system operators, etc.