SMART HEALTHCARE INFRASTRUCTURE: INTEGRATING AI, SUSTAINABLE ENGINEERING, LEGAL COMPLIANCE, AND BUSINESS MODELS FOR AFFORDABLE PHARMACEUTICAL DISTRIBUTION

Abstract

The rise of fairness in the provision of healthcare and sustainability of delivery has led to the demand for applicable, intelligent and low-cost solutions. In the thesis, a multi-disciplinary smart healthcare infrastructure was developed that applied artificial intelligence, sustainable supply chain optimization, legal compliance frameworks and adaptive business models to improve pharmaceutical access. A dataset of pharmaceutical sales, compliance reports and regional demand statistics were used to implement four algorithms. Random Forest (RF) was used for demand forecasting; genetic algorithms (GA) were used for sustainable supply chain optimization; support vector machine (SVM) was used to monitor compliance; and reinforcement learning (RL) was used to manage adaptive pricing and business models. The experimental a analyses show that RF has a precision of 93.6% in predicting region demand, while GA delivered a reduction of logistics-costs of 28.4% compared to comparison models. SVM delivered a precision of 91.2% in classifying compliance violations; and RL reduced the end user price, as an affordability metric, by an average of 17.5%. In relation to existing works in the domain and category, the performance of the models and algorithms were superior in predicting reliability and cost performance. This thesis presented the technological, regulatory, and economic aspects required to develop a sustainability model explaining healthcare delivery based on compliance sustainability. Based on this aspect it was determined that multi-disciplinary approaches that use intellect could enhance the allocation of pharmaceutical elements by balancing affordability, sustainability and compliance.