In the transition to intelligent systems, lean production faces challenges from high-mix, low-volume manufacturing and frequent disturbances.Traditional resource allocation methods are inadequate for these dynamic environments.Deep Reinforcement Learning (DRL) offers a solution but often lacks alignment with lean principles and is hard to interpret, limiting effective human-machine collaboration.This study introduces an interpretable DRL approach for dynamic resource allocation in lean production.Utilizing a Markov decision process, the model includes a state space with resource efficiency, system balance, and inventory status, and a reward function targeting waste reduction and efficiency.The approach uses a Temporal Convolutional Network (TCN) to capture temporal dependencies and employs visualization tools like Global Average Pooling (GAP) and Class Activation Mapping (CAM) to trace decisions to specific waste elimination goals.This integrated learning process enables real-time decision verification by on-site managers, offering a practical solution for complex production settings and enhancing integration of DRL with lean management.

This study applied the design of experiment (DoE) to a hydraulic engine mount.Taguchi L16 orthogonal array was selected as an experimental plan, and various mounts were assembled and tested.The dynamic stiffness, Kd, values at 50 Hz frequency for 0.1 mm vibration amplitude as the reference value were concluded.The best design combination, the priority order, and the contribution ratios of design parameters were determined by applying signal-to-noise ratio (S/N) and variance analysis (ANOVA).In the analyses, it was found that dynamic hardening, , and fluid volume, V, were much more effective than other parameters on the Kd.Then, the regression analysis was applied to create two equations considering all parameters and only two effective parameters.It was obtained that Kd values calculated from the two equations were close to each other and in good agreement with the experimental results of the mounts in the DoE and the other two mounts not included in the DoE.This study has shown that analytical equations that give accurate results can be concluded for Kd via the Taguchi method.

Autonomous robots are of utmost importance across various research fields and have numerous practical applications. They can be used in scenarios such as rescue operations, security, safety management, and interventions in hazardous or hostile environments, among others. A key application of these robots is overcoming obstacles in extreme conditions, especially when human intervention is either impossible or not permitted. This paper presents modelling and simulation results based on a mathematical tool developed through systematic data acquisition and analysis, ensuring the safe operation of a robot designed to assist its pilot. The proposed predictive model is built using data collected from the robot's sensors, which are stored in a dedicated database. This data includes information on position, velocity, and acceleration, as well as their handling and processing in the presence of obstacles. The paper also includes a case study, with results that can serve as a predictive model applicable to similar conditions.

With global economic integration and rapid IT advancements, enterprises face complex market environments and intense competition.Efficient management and coordination of supply chains and manufacturing are critical.This study explores optimization strategies for integrated supply chain and manufacturing management using system dynamics.Initially, it investigates the contractual collaboration model based on system dynamics, analysing dynamic behaviours and optimization strategies.Subsequently, a system dynamics simulation model of the integrated system is constructed.Simulation analysis reveals the interactive relationships and dynamic patterns within the system.While many studies focus on static analysis, they often overlook dynamic changes and complex interactions.This study enriches the theoretical framework and provides solutions for enhancing management efficiency and competitiveness in a volatile market.

While large enterprises are actively preparing for digital transformation by leveraging technologies such as digital twins, smaller companies face challenges due to economic constraints and market uncertainties, leading to a relative lack of awareness and readiness.To address this situation, this study proposes a digital twin development framework tailored for small and medium-sized enterprises (SMEs).This framework utilizes attachable IoT devices for real-time collection of manufacturing data and leverages public server systems for data management.Moreover, it enables the refinement and optimization of digital twins by training machine learning models on collected data.Additionally, the framework includes the integration of simulation models and machine learning models for comprehensive digital twin modelling.Finally, the paper suggests a process for applying and validating this framework in real manufacturing companies, demonstrating the effects of digital twin implementation on productivity enhancement in the production lines of two SMEs.

[EN] Freight transport is a fundamental activity for the economic growth of a country or region, and the transport sector is one of the main contributors to gross domestic product (GDP) measurements. The total tons to be transported per year and transport types must be taken into account by state organisations for infrastructure planning, routes, prices and taxes. Governments make investment decisions according to the size of the freight mobilised per year. Transportation companies should contemplate external and internal variables, for example, fuel prices, distances, environmental impact, among others. Sustainability has become a determining factor in transport companies' decision making. This paper presents a novel study about the factors that affect freight land transport from the sustainability perspective. Then it proposes a conceptual framework to act as a roadmad to build a simulation model of freight land transport by defining the key parameters in economic, social and environmental terms. (Received in December 2023, accepted in August 2024. This paper was with the authors 5 months for 2 revisions.)

The presented article deals with the issue of inventory management optimization through modelling and simulation in the software Tecnomatix Plant Simulation.Based on an in-depth analysis of the supply chain with an emphasis on inventory management, a simulation model was developed.It was optimized using the inventory theory methodology in combination with a discrete-event simulation approach.Dynamic simulation made it possible to perform experiments and optimize the existing state.The solution to the problem is elaborated on a case study to achieve the efficiency of the supply process of a specific operation of a clothing retail chain, which is struggling with the problems of excessive and inaccurate supply.