This dissertation deals with applications of Prolog and Artificial Intelligence (AI) to chemical engineering, and in particular, to the area of chemical process synthesis. We introduce the language Prolog (chapters 1-9), discuss AI techniques (chapters 10-11), discuss EXSEP, the EXpert System for SEParation Synthesis (chapters 12-15), and summarize applications of both AI and Artificial Neural Networks (ANNs) to chemical engineering (chapters 16-17). We have developed EXSEP, a knowledge-based system that performs separation process synthesis. EXSEP is a computer-aided design tool that can generate flowsheets using any combination of high-recovery (sharp) and low-recovery (nonsharp) separations, using a variety of separation methods with energy and mass separating agents. EXSEP generates separation process flowsheets using a unique plan-generate-test approach that incorporates computer-aided tools and techniques for problem representation and simplification, feasibility analysis of separation tasks, and heuristic synthesis and evolutionary improvement. A difficult problem in knowledge-based approaches to chemical engineering is the "quantitative or deep knowledge dilemma." Experience has shown that a strictly qualitative knowledge approach to chemical process synthesis is insufficient. However, including rigorous quantitative analysis into an expert system is cumbersome and impractical. EXSEP overcomes this deep-knowledge dilemma through a unique knowledge representation and problem-solving strategy that includes shortcut design calculations. These calculations are used as a feasibility test for all separations; no separation is chosen by EXSEP unless it is deemed as thermodynamically feasible through this quantitative, deep-knowledge, engineering analysis. We apply EXSEP for the flowsheet synthesis of several industrial separations problems. The results show that EXSEP successfully generates technically feasible and economically attractive process flowsheets accurately and efficiently. EXSEP is also user-friendly, and can be readily applied by practicing engineers using a personal computer. In addition, EXSEP is developed modularly, and can be easily expanded in the future to include additional separation methods. / Ph. D.