A Supply Chain Ecosystem Approach to Smart City Manufacturing Clusters: Digital Transformation-Enabled Energy Management in Urban Water Systems

This study develops a supply chain ecosystem (SCE) approach for smart city manufacturing clusters (SCMCs) through the integration of digital technologies into energy management of urban water systems (UWSs). The proposed framework integrates the Internet of Things (IoT), blockchain, and machine learning (ML) analytics within a smart water grid architecture to support sustainable and energy-efficient water supply management. Its principal contribution is an SCE-oriented information and analytical framework linking distributed system entities and stakeholders through integrated data management and decision support. To address the hierarchical and multi-objective nature of the energy management problem, a stratified representation approach is proposed, enabling decomposition into interconnected subproblems and improving coordination across multiple system levels within the SCE. An information-analytical system is developed, integrating subsystems for data acquisition, processing, storage, analysis, planning, monitoring, and decision support. The proposed architecture combines IoT-enabled real-time data acquisition with blockchain-based mechanisms supporting data integrity, traceability, and verifiable information exchange among system entities. Energy planning is formalised through the construction of an energy baseline (EnB) accounting for system hierarchy, operational constraints, and planning horizon. The operational application of the framework is demonstrated using a case study of a water intake facility. GMDH (Group Method of Data Handling) neural networks are applied to model energy consumption, while multi-criteria analysis identifies a second-order polynomial activation function as optimal, achieving high predictive accuracy (MAPE = 0.513%; RMSPE = 0.595%). The resulting day-ahead EnB, based on planned technological parameters and pump operation schedules, demonstrates how the proposed SCE framework translates integrated data management into operational decision support for energy-efficient UWS operation.