A study on the genetic algorithm for the scheduling problem in cell culturing
세포배양 일정계획 문제를 위한 유전알고리즘에 관한 연구
- 주제(키워드) 도움말 유전알고리즘 , 일정계획
- 발행기관 강릉원주대학교 산업대학원
- 지도교수 도움말 성기석
- 발행년도 2024
- 학위수여년월 2024. 8
- 학위명 석사
- 학과 및 전공 도움말 산업대학원 산업공학과
- 실제URI http://www.dcollection.net/handler/kangnung/000000011820
- UCI I804:42001-000000011820
- 본문언어 영어
초록/요약 도움말
This research explores innovative approaches to optimize scheduling complexities inherent in cell culturing, a critical aspect of biopharmaceutical manufacturing. Drawing insights from seminal works in scheduling problems, the study introduces a tailored genetic algorithm (GA) framework designed specifically for cell culturing scheduling. Key enhancements to the GA framework include reverse chromosome design, fitness scaling, specialized repair mechanisms, and a mixed mutation strategy. These modifications aim to improve solution diversity, adaptability, and efficiency in addressing the unique challenges of cell culturing scheduling. Experimental results demonstrate the robust performance of the proposed GA framework across various planning horizons and culturing durations. Notably, the mixed mutation strategy plays a pivotal role in efficiently exploring the solution space, balancing exploration and exploitation to discover high-quality solutions within complex constraints. The study underscores the importance of mutation operators in algorithmic design for scheduling problems, particularly in addressing constraint violations introduced during crossover and mutation operations. Overall, this research contributes valuable insights to scheduling optimization for cell culturing applications, highlighting the significance of adaptive algorithmic approaches in addressing real-world challenges in biopharmaceutical manufacturing.
more목차 도움말
I. INTRODUCTION 1
II. LITERATURE REVIEW 2
2.1 Conceptual framework 2
2.1.1 Cell culture 2
2.1.1.1 Growing process 3
2.1.1.2 Equipment 4
2.1.1.3 Cross-contamination 5
2.1.2 Genetic Algorithm (GA) 6
2.1.2.1 Initial population 8
2.1.2.2 Selection 9
2.1.2.3 Crossover 10
2.1.2.4 Mutation 12
2.2 State of the art 13
2.2.1 Resource constrained project scheduling problem 13
2.2.2 Optimization model for scheduling in biopharmaceutical manufacturing 13
2.2.3 GA approach for biopharmaceutical manufacturing capacity planning and scheduling 14
III. METHODOLOGY 16
3.1 Problem formulation 16
3.2 Initial approaches and evolution 18
3.3 Genetic Algorithm implementation 19
3.3.1 Chromosome design and initial population generation 21
3.3.2 Fitness scaling and selection 23
3.3.3 Repair mechanism 24
3.3.4 Mixed mutation 25
3.4 Parameter settings and experimental setup 27
IV. RESULTS 28
V. CONCLUSIONS 31
VI. REFERENCES 33
Abstract 38
