The pharmaceutical supply chain plays a crucial role in ensuring timely and efficient delivery of medicines, yet it faces persistent challenges related to waste generation, overproduction, and resource inefficiency. These inefficiencies contribute not only to economic losses but also to environmental and ethical concerns surrounding the disposal of expired and unused medical products. This research proposes a Cognitive Decision Intelligence (CDI) framework that integrates Natural Language Processing (NLP) and Big Data Analytics to achieve sustainable pharmaceutical supply chain optimization with a focus on medical waste prevention. The proposed approach utilizes linguistic data interpretation, real-time data streams, and predictive analytics to support intelligent, data-driven decision-making. By merging NLP-based semantic understanding with machine learning-driven forecasting, the CDI system is capable of predicting demand, identifying waste risks, and recommending corrective measures. Experimental simulations demonstrate that the proposed model can significantly reduce inventory redundancy and improve sustainability metrics. Furthermore, ethical and explainability considerations are embedded within the design to ensure transparency and accountability. The findings underscore how cognitive analytics can transform traditional pharmaceutical logistics into an adaptive, sustainable, and intelligent ecosystem. The inclusion of blockchain ensures data authenticity and ethical transparency in supply chain operations.