Industry 4.0 solutions for optimised value recovery from Waste Electrical and Electric Equipment (WEEE)
Demanufacturing activities consist of the disassembly of products to their separate components for recovery or further recycling as well as the disassembly or destructive processing of the separate components for the recovery of valuable materials. Remanufacturing activities consist of the subsequent cleaning, repairing, as well as refurbishing, upgrading and maintenance of recovered components or entire product assemblies. The potential of de- and remanufacturing approaches is evidenced by multiple research results. For example, demanufacturing flat screen television sets for optimal purity material recycling leads to >90% of precious metal recovery and >50% of technical plastics recycling, as opposed to <10% precious metals and 15% technical plastics recycling in shredder based scenarios . When additionally re-manufacturing is also considered, the potential further increases, since the functional value of products can reach up to 500 times the material value, for example for smartphones. If the average residual value of a component is, for example, 100 times the material value, then recovering 1 component out of 100 results in doubling the achievable revenue! Even though Flanders has one of the most advanced networks for reuse by, among other companies and organizations, the Kringwinkels, the reuse potential is far from being fully exploited. In contrast, a decrease in reuse activities is feared as a consequence of the increasing product complexity and, accordingly, required expertise, whereas already today it is impossible to meet the increasing demand for reused products. Therefore, there is a strong need for smart systems that can identify products or components to facilitate decision making during demanufacturing activities to determine optimal value recovery routes for the product at hand.
Prior research on value recovery systems has focused on automation of both non-destructive disassembly as well as destructive dismantling processes for, among others, washing machines , , personal computers , LCD monitors –, mobile phones , , remote controls , digital cameras  and End-of-Life Vehicles (ELVs) –. A notable research project by electronics manufacturer Apple Inc in 2016, called “Liam”, yielded a system responsible for the automatic disassembly of one of its own products, the iPhone 6, with as main goal to separate the components for material recycling. Liam comprises of 29 robotic cells and is capable of disassembling 1.2 million iPhone 6 units a year into 8 distinct components . However, this system is designed to disassemble one specific product model only, and is, according to Apple, only used to disassemble defective units into components for material recycling. Moreover, the specific business case of high-volume series, short-product life time, high material values, and established product return policies allows a centralized and relatively straightforward automation approach. For most products, however, the end-of-life chain is characterized by an extremely large variety of product types, brands and models that are collected over a larger time span. Costs of reverse-logistics associated with a dedicated centralized system would moreover be excessive, and only regional (500km – 1000km) recollection and processing is realistic. Therefore, the major challenge for the future circular economy is to realize systems that are as performant as the Liam system in terms of value recovery, yet can deal with a very large variety of product models and brands. The flexibility and intelligence required for a demanufacturing system hence inherently bears the same requirements as the Industry 4.0 paradigm puts forward for the manufacturing industry, i.e. dealing with sensor systems, big data, and real-time analytics . Smart product and component recognition, complemented with real-time decision making on optimal value recovery strategies for the product at hand is therefore envisaged in this research. Since availability of product data (such as material composition) from the original producer cannot be guaranteed – an issue ignored in state-of-the-art algorithms for the determination of optimized disassembly strategies  – strategies for self-learning of the envisaged de-manufacturing solution are also required.
The overall aim of this PhD research project is to conceive and investigate the feasibility of realizing smart demanufacturing solutions that enable optimal resource recovery in a circular economy. Therefore, it envisages the translation of the Industry 4.0 paradigm from the manufacturing sector towards the de-manufacturing sector. Such systems can provide solutions regarding the challenges in handling high product variety and the increasing product complexity, with flexible automated de- and re-manufacturing systems with integrated decision making support. This goal will be pursued by realizing the following technical objectives.
Development of a method that enables autonomous yet robust identification of WEEE products. The project will therefore investigate and test the feasibility and potential of computer vision systems and associated algorithms for decision making under uncertainty for the recognition of products and components.
Development of an autonomous decision making system for value recovery strategy determination for detected products. A decision making system is envisaged that determines the optimal value recovery strategy for each identified product. Based on continuously updated market prices and storage space information, the algorithms will determine which components to disassemble for reuse or remanufacture, and how to dismantle the remaining structure for optimal resource and value recovery.
Development of a Product Data Management system for the envisaged Demanufacturing 4.0 solution. The extreme diversity of product data required for the demanufacturing 4.0 solutions calls for well-structured and efficient databases. Starting from the available expertise within the Life Cycle Engineering group, the research will explore which data is necessary for de- and re-manufacturing activities, and will propose a well-structured database system for the storage of valuable product data and dismantling instructions. Feasibility of connecting to external databases will be investigated and the logging of scanned products by the computer vision system will be incorporated. Identifying an optimal structure for the database is an important technical objective as this will influence the performance of searching and storing data.
Realizing self-learning capabilities in demanufacturing systems. The self-learning character of the system allows for expanding or updating and improving the database when recognition is not possible due to a low percentage of matched data or insufficiently recognised features. This way, the system improves itself by increasing the accuracy of the available data which improves future recognition.
Validation of the conceived solutions. A well selected product category (e.g. laptops) will be used to validate the proposed algorithms.
 J. R. Peeters, P. Vanegas, J. R. Duflou, T. Mizuno, S. Fukushige, and Y. Umeda, “Effects of boundary conditions on the end-of-life treatment of LCD TVs,” CIRP Ann.-Manuf. Technol., vol. 62, no. 1, pp. 35–38, 2013.
 G. Seliger, B. Basdere, T. Keil, and U. Rebafka, “Innovative Processes and Tools for Disassembly,” CIRP Ann. - Manuf. Technol., vol. 51, no. 1, pp. 37–40, 2002.
 J. R. Duflou, G. Seliger, S. Kara, Y. Umeda, A. Ometto, and B. Willems, “Efficiency and feasibility of product disassembly: A case-based study,” CIRP Ann. - Manuf. Technol., vol. 57, no. 2, pp. 583–600, 2008.
 F. Torres, P. Gil, S. T. Puente, J. Pomares, and R. Aracil, “Automatic PC disassembly for component recovery,” Int. J. Adv. Manuf. Technol., vol. 23, no. 1–2, pp. 39–46, 2004.
 H.-J. Kim, S. Kernbaum, and G. Seliger, “Emulation-based control of a disassembly system for LCD monitors,” Int. J. Adv. Manuf. Technol., vol. 40, no. 3, pp. 383–392, Jan. 2009.
 S. Vongbunyong and W. H. Chen, Disassembly Automation. Sidney, Australia: Springer International Publishing, 2015.
 S. Vongbunyong, S. Kara, and M. Pagnucco, “Learning and Revision in Cognitive Robotics Disassembly Automation,” Robot Comput-Integr Manuf, vol. 34, no. C, pp. 79–94, Aug. 2015.
 S. Vongbunyong, S. Kara, and M. Pagnucco, “Basic behaviour control of the vision-based cognitive robotic disassembly automation,” Assem. Autom., vol. 33, no. 1, pp. 38–56, 2013.
 B. Basdere and G. Seliger, “Disassembly Factories for Electrical and Electronic Products To Recover Resources in Product and Material Cycles,” Environ. Sci. Technol., vol. 37, no. 23, pp. 5354–5362, Dec. 2003.
 P. Kopacek and B. Kopacek, “Intelligent, flexible disassembly,” Int. J. Adv. Manuf. Technol., vol. 30, no. 5, pp. 554–560, Sep. 2006.
 P. Schumacher and M. Jouaneh, “A system for automated disassembly of snap-fit covers,” Int. J. Adv. Manuf. Technol., vol. 69, no. 9, pp. 2055–2069, Dec. 2013.
 M. Merdan, W. Lepuschitz, T. Meurer, and M. Vincze, “Towards ontology-based automated disassembly systems,” in IECON 2010-36th Annual Conference on IEEE Industrial Electronics Society, 2010, pp. 1392–1397.
 J. Li, M. Barwood, and S. Rahimifard, “An automated approach for disassembly and recycling of Electric Vehicle components,” in 2014 IEEE International Electric Vehicle Conference (IEVC), 2014, pp. 1–6.
 A. Sánchez et al., “Automatic Disassembly System Architecture for End-of-life Vehicles,” in Proceedings of the 9th WSEAS International Conference on International Conference on Automation and Information, Stevens Point, Wisconsin, USA, 2008, pp. 68–73.
 A. J. Sánchez-Salmerón, M. Mellado, C. Ricolfe, A. Valera, J. J. Esteve-Taboada, and M. Giménez, “VIRTUAL PLATFORM FOR PROTOTYPE IMPLEMENTATION OF FLEXIBLE AUTOMATED DISASSEMBLY SYSTEMS,” IFAC Proc. Vol., vol. 40, no. 2, pp. 87–95, Jan. 2007.
 C. Rujanavech, J. Lessard, S. Chandler, S. Shannon, J. Dahmus, and R. Guzzo, “Liam - An Innovation Story.” Sep-2016.
 pwc, “Industry 4.0: Building the digital enterprise - 2016 Global Industry 4.0 Survey - Industry key findings.” pwc, 2016.
 A. K. Parlikad and D. McFarlane, “RFID-based product information in end-of-life decision making,” Control Eng. Pract., vol. 15, no. 11, pp. 1348–1363, 2007.