Automated Modelling, Scheduling and Control for Manufacturing as a Service

The manufacturing landscape consists of large-scale manufacturing systems that become available, or unavailable, on-the-fly and can be accessed remotely. These systems consist of ‘assets’, such as raw materials, robots, software, hardware, parts of or whole factory floors, transportation fleets, logistic routes, that can be connected to each other with different possible configurations. Currently, supply chains are usually formed at the beginning of a product or service cycle, are static, described by fragmented models of different types. Commonly, supply chain forming and optimisation depends on teams of experts deciding the allocation of resources and the scheduling of different components forming the chain.

Manufacturing as a Service is a new paradigm, aiming to connect in real-time manufacturing resources, and automate scheduling and decision-making by creating flexible and decentralised supply and value chains. Our research, aligned with efforts of a wider consortium of research institutes and manufacturers, will focus on addressing the challenges related to this new type of operation from a control theoretic perspective, especially in modelling, scheduling and resource allocation of supply chains and service chains. We will utilise tools from control theory, optimization and computer science, starting from recent advances in hybrid systems theory, set-based methods and reachability analysis, learning, data-driven methods, and formal methods.

Project Description:

Our first goal is to explore new modelling structures for predicting, controlling, and optimising supply chains that are subject to abrupt changes and may switch on demand or by change of operating conditions. We will focus on models valid for a range of possible parameters, incorporating volatility of demand and time varying availability of resources. Dynamics are event-driven and switching, capturing the versatile allocation of tasks. We will start from earlier work and structures taken, e.g., from queuing theory and discrete event systems, as well as hybrid systems, namely switching linear systems, timed automata with rectangular/zonotopic constraints. Our second goal is to develop methods for analysis of key performance indicators, translated as state variables in the models generated. We will use computation-aware tools, based on reachability analysis, set-based and data-driven methods. The third goal is to combine control-theoretic, graph-theoretic, data-driven optimisation tools as well as tools from formal methods, and create feedback decision mechanisms for control of manufacturing resources (digital, physical, etc.) for complex specifications: Our decision variables are both logical and continuous: binary decisions reconfigure manufacturing modules, choice of providers, logistic routes etc., and continuous decisions act on request flows, allocation and duration of time slots, workforce, etc.

Commencement of the PhD project aligns with the start of a 36-month EU-funded 6.5 million EUR research project on enabling the concept of ‘Manufacturing as a Service’. Our consortium consists of 18 partners, including seven internationally leading universities in Europe and Canada, Manufacturing companies and research centres with complementing expertise. There will be multiple opportunities for international collaboration, such as research visits, joint research work etc.

For queries contact Nikolaos Athanasopoulos [email protected]

• PhD Start Date: Expected 1 January 2025.
• Application closing date: 31 August 2024.

Academic Requirements:

The minimum academic requirement for admission is normally an Upper Second Class Honours degree from a UK or ROI Higher Education provider in a relevant discipline, or an equivalent qualification acceptable to the University.

Candidates should have a background in Engineering, Computer Science, or Applied Mathematics, preferably trained in Control theory / Control engineering.

To apply, please complete an application through the Direct Applications Portal: https://dap.qub.ac.uk/portal/user/u_login.php