Artificial Intelligence (AI) assisted Industrial Symbiosis (IS)

“产业共生（IS）被定义为两个或多个产业之间通过交换/共享材料、能源、服务和/或知识来发展互利关系。在过去几年中，由于一些障碍，主要是财政和社会障碍，这些计划的发展已经停止。信息系统从业者一致认为，需要能够帮助克服这些障碍的促进者。计算算法及其通过信息和通信技术（ICT）工具的应用可以发挥这一作用。与此同时，工业生态系统的日益数字化（工业4.0）和物联网（IoT）网络的广泛部署，导致了大量数据的产生和捕获。人工智能可以提供广泛的强大技术，可以通过处理获取的数据和支持管理IS生态系统的复杂和动态方面来提供智能。然而，只有有限数量的开发工具利用人工智能工具来解决与IS开发相关的问题，主要集中在与工业设施、废物生产和供应链经济学相关的数据分析上。该项目的目标是通过机会识别、效率评估和控制，以及协助相关利益相关者的决策过程，为动态优化IS奠定理论基础并开发AI框架，以此作为增强IS生态系统基础的基础。先前开发的一种工具，已开发用于促进IS计划（基于固体，液体和气体废物流），将通过利用人工智能领域的最新进展来改进和扩展。现有的工作证明了一组机器学习技术用于识别IS机会的适用性，包括推荐算法，如关联规则挖掘、基于案例的推理、协同过滤、基于知识的推荐和基于规则的推荐。开发的算法将能够：(i)探测有利的地理区域和工业部门以建立信息系统计划；检测需要处理的物料、废物和能源流动中的异常情况，以优化处理过程；（iii）分析及预测可能影响资源供求的重大事件，为预测供求平衡及优化物流提供基础；支持根据用户自定义偏好优化IS匹配；(v)通过考虑与已纳入计划的利益攸关方的共生关系，自动确定供信息系统计划考虑的其他适当废物，并建议一个优化的价值链。”

"Industrial Symbiosis (IS) is defined as the development of mutually beneficial relationships between two or more industries, by exchanging/sharing material, energy, services and/or knowledge. Over the last years, the development of such schemes has been halted due to several barriers, mostly financial and social. IS practitioners have agreed that there is need for facilitators that can help overcome those barriers. Computational algorithms, and their application through Information and Communication Technology (ICT) tools, can play this role.At the same time, the increasing digitalization of industrial ecosystems (Industry 4.0) and the widespread deployment of Internet of Things (IoT) networks, leads to the generation and capture of huge amounts of data. Artificial intelligence can provide a wide range of robust technologies that can deliver intelligence through processing the acquired data and supporting the management of complex and dynamic aspects of IS ecosystems. However, only a limited number of the developed tools have exploited AI tools to address problems related to IS development, mostly focusing on data analysis related to industrial facilities, waste production and supply chain economics.The objective of this project will be to lay the theoretical foundations and develop an AI framework for the dynamic IS optimization, through opportunity identification, efficiency assessment and control, and assisting in the decision process of the involved stakeholders as the basis towards enhancing the foundation of IS ecosystems. A previously developed tool, which has been developed for the facilitation of IS schemes (based on solid, liquid and gaseous waste streams), will be improved and extended by leveraging on recent advances in the artificial intelligence field. Existing work demonstrated the suitability of a set of machine learning techniques to identify IS opportunities, including recommender algorithms such as association rule mining, case-based reasoning, collaborative filtering, knowledge-based recommendation, and rule-based recommendation.The developed algorithm will be able to: (i) Detect favourable geographic areas and industrial sectors to establish IS schemes; (ii) Detect anomalies in the flow of materials, waste and energy that need to be treated to optimise the process; (iii) Analyse and predict significant events that may affect the demand and supply of resources providing the basis for predictive demand-supply balancing and logistics optimisation; (iv) Support the optimisation of IS matches based on user-defined preferences; (v) Automatically identify additional suitable waste products to be considered by IS schemes, and suggest an optimized value chain, by considering symbiosis with stakeholders already included in the schemes."