Digital solutions have become increasingly important for manufacturing companies to increase their productivity, effectiveness, and competitiveness in a global market, which demands low prices, high quality, and fast delivery times. In order to improve production efficiency, it is also necessary to optimize transportation activities in the production floor via digitization and automation of those processes. Many companies have already used or are planning to use autonomous mobile robots (AMR) to manage production logistics more effectively. The rapid development of the Internet of Things (IoT) and the advanced hardware and software of AMR allow them to perform autonomous tasks in dynamic environments, where they can communicate and independently coordinate with other resources, such as machines and systems, and thus decentralize the decision­-making steps of manufacturing processes. Decentralized decision making allows the manufacturing system to dynamically adapt to changes in the system state and environment. Such developments have affected traditional planning and control methods and decision­-making processes, but they also require the software and embedded artificial intelligence (AI) algorithms to be more capable of executing these decisions. In this study, we describe how to use a 3D virtual factory concept to integrate an AMR system with AI functionality into the production logistics of the food industry. The paper presents an approach to analyze the performance of AMR in the transportation of goods on the manufacturing plant floor, based on the creation and simulation of the 3D layout, the monitoring of key performance indicators (KPI), and the use of AI for proactive decision making in production planning. A case study of the food industry demonstrates the relevance and feasibility of the proposed approach.

Modern manufacturing systems operate in a global and competitive environment, where sustainability has become a critical driver for performance. Performance measurement, as a method for monitoring enterprise processes, plays a central role in aligning operational efficiency with sustainable development goals. Recently, a number of different frameworks, systems, and methods have been proposed for small and medium enterprises. Key performance indicators (KPIs) are known to be powerful tools which provide accurate information regarding bottlenecks and weak spots in companies. The purpose of the current study is to develop an advanced KPI selection/prioritization model and apply it in practice. The initial set of KPIs are obtained based on a literature review. The expert’s knowledge, outlier methods, and optimization of the enterprise analysis model (EAM) are utilized for reducing the initial set of KPIs. A fuzzy analytical hierarchy process (AHP) is implemented for prioritization of the criteria. Five different MCDM (multi-criteria decision-making) algorithms are implemented for prioritization of the KPIs. The recently introduced RADAR method is extended to the fuzzy RADAR method, providing a flexible approach for handling uncertainties. An analysis and comparison of the rankings obtained by utilizing five MCDM algorithms is performed. The prioritized KPIs provide valuable input for improving KPIs with the highest impact in particular small and medium-sized enterprises (SMEs) when implementing sustainability-aligned performance metrics.

Recent trends in manufacturing such as Industry 4.0 have brought the researchers' attention to smart production intralogistics. AGVs, especially mobile robots play a vital role in this development. Contrarily, industrial internet technologies offered new possibilities for information exchange,...

Small and medium-sized enterprises (SMEs) in the manufacturing sector often struggle to make effective use of production data due to fragmented systems and limited digital infrastructure. This paper presents a case study of implementing an AI-enhanced digital twin in a custom sportswear manufacturing SME developed under the AI and Robotics Estonia (AIRE) initiative. The solution integrates real-time production data collection using the Digital Manufacturing Support Application (DIMUSA); data processing and control; clustering-based data analysis; and virtual simulation for evaluating improvement scenarios. The framework was applied in a live production environment to analyze workstation-level performance, identify recurring bottlenecks, and provide interpretable visual insights for decision-makers. K-means clustering and DBSCAN were used to group operational states and detect process anomalies, while simulation was employed to model production flow and assess potential interventions. The results demonstrate how even a lightweight AI-driven system can support human-centered decision-making, improve process transparency, and serve as a scalable foundation for Industry 5.0-aligned digital transformation in SMEs.

From smart factories to service applications, human-robot interaction is crucial to the development of collaborative settings in which humans and robots operate side by side. Since robots in shared spaces must take into consideration variables such as human motion unpredictability and potential workplace risks, effective risk management is essential. When building these systems, striking a balance between factors such as safety, ergonomics, and operational flexibility becomes crucial. Misalignment between human purpose and robot behaviors might result in accidents or increased injury risks, especially in environments with high physical demands. One of the key challenges in human-robot interaction is handling uncertainty. The main aim of the current study is to introduce a conceptual framework for safety/risk analysis, including a hierarchy tree of the risk criteria and risks. Both human- and robot-related factors are considered. The multi-criteria decision-making procedure developed for autonomous vehicle systems is adapted for risk analysis in human-robot interaction. As a final result, the prioritized risk criteria and risks are identified. These results lay the foundation for reducing risks in the future.

This paper examines the development of a virtual factory model to optimize overall equipment effectiveness (OEE) in a planned manufacturing facility. Using digital simulations based on a wood manufacturing setup, AI-driven models can be applied to analyze specific OEE metrics, allowing for targeted identification of production bottlenecks and efficiency improvements. The virtual factory enabled scenario testing for the proposed facility, providing actionable insights without impacting current operations. The preliminary results indicate that AI integration within a virtual factory can significantly enhance planning and decision-making for future production investments.