DC7 Project: Explainable Artificial Intelligence for internal logistics systems in green manufacturing

Doctoral Candidate

Myroslav Mishchuk, MSc

Myroslav Mishchuk received his Master’s degree in Computer Science with Honors from Lviv Polytechnic National University. His academic background includes Artificial Intelligence, Machine Learning, Data Analysis, Neural Networks, Microcontroller Programming, and Digital Electronics. His research expertise spans areas such as Human Activity Recognition (HAR), Biomedical Signal Processing, Industry 4.0 and 5.0, Machine Learning, and Neural Networks, closely aligning with the goals of the TUAI project: Explainable AI (XAI) for internal logistics and AGVs.

Myroslav_Mishchuk_Self-presentaion.pdf

Main Supervisor: Rafał Cupek (SUT) 

Co-Supervisors: David Camacho (UPM), Salvatore Cuomo (UNINA), Jia-Chun Lin (NTNU) 

R&D cooperation: AIUT

Objectives: develop novel methodologies for internal logistics systems based on AGVs, with a focus on energy- and time-efficient planning, scheduling, and analysis. 

The eXplainable AI (XAI) approach will be applied to make the decision-making processes of 'black box' AI solutions more transparent and modifiable for human users, without sacrificing predictive accuracy. XAI will serve as a tool to ensure that similar instances lead to similar explanations, and different instances lead to different explanations, while also highlighting any inherent biases in the data. Measures of similarity and separability are key to evaluating these aspects and ensuring that the explanations are both meaningful and trustworthy.

Research challenges include addressing potential bias and unfairness, ensuring that AI systems can clearly articulate their decisions, and developing AI that can communicate effectively with humans. 

Expected Results: include a fully integrated production environment with active, real-time interactions among all related elements in the ecosystem. The XAI-based services will support human–machine communication (e.g., between operators and machines), warehouse and logistics management software, collaborative robots, Manufacturing Execution Systems, and other actors involved in a green manufacturing environment.

Applied research: will focus on novel methods for energy management and online process data monitoring, enabling deeper knowledge acquisition of intralogistics processes. Additionally, methods for agile communication between intralogistics components and third-party systems have the potential to expand the portfolio of technologies used by AIUT.

Planned secondment(s): UPM(4 months); UNINA(4 months); NTNU (4 months)                   

Enrolment in Doctoral degree: SUT