隨著數位轉型與人工智慧技術發展，製造現場逐漸朝向以資料輔助決策之管理模式。然而，製藥產業之品質檢驗流程受法規遵循、品質追溯與交期要求影響，排程需同時考量產品類型、檢驗項目、人員能力、設備限制與工作時段等條件。若資料分散於不同系統或人工紀錄中，排程決策容易依賴個人經驗，面對急件插單、人員異動或設備異常時，亦較難快速形成一致且可驗證的調整依據。
本研究以製藥品質檢驗流程為對象，提出品質檢驗排程架構。首先，以自働生產管理環（Jidoka–JIT Cycle, JJC）作為研究流程基礎，並透過資訊流圖（Information Stream Mapping, iSM）分析現行流程之資訊傳遞問題。接著，運用低程式碼平台整合批次資料與檢驗規格限制，建立可供排程演算法使用之結構化資料。在排程求解方面，本研究採用模擬退火法建立排程模型，並以實驗設計與收斂趨勢輔助確認參數設定。為提升突發事件下之調整能力，本文進一步比較兩種生成式 AI 介入模式，包括直接產生調整排程，以及行事件解析後，再由模擬退火法執行重排。
實證結果顯示，在系統導入方面，資訊流自動化與即時化程度皆由 0% 提升至 80%，顯示低程式碼平台有助於改善資料分散與人工傳遞問題。在排程績效方面，模擬退火法將平均檢驗 Lead Time 由人工排程之 171.0 小時降至 64.2 小時，改善約 62.5%，並優於最早交期優先規則之 122.7 小時。在 AI 輔助調整方面，相較於大型語言模型直接產生排程，事件解析結合模擬退火重排之 Model II 於測試情境中維持 100% 可執行率與 0% 幻覺率，平均運行時間亦由 184.4 秒降至 37.4 秒。本研究整合資訊流程分析、系統建構、排程求解與AI事件調整流程之排程架構，可作為製造現場導入排程與決策支援系統之參考。

Quality inspection plays a critical role in pharmaceutical manufacturing, as products must pass multiple inspection procedures before release. However, inspection scheduling is often affected by product types, test items, staff skills, equipment availability, working hours, and due-date requirements. In practice, when inspection-related data are scattered across different systems or manual records, scheduling decisions tend to rely heavily on personal experience. This may lead to inconsistent task sequences and limited ability to respond to urgent orders, staff changes, or equipment downtime.
This study develops a digital-lean-oriented quality inspection scheduling system and examines the role of generative AI in dynamic schedule adjustment. The proposed framework is based on the Jidoka–JIT Cycle (JJC), using Information Stream Mapping (iSM) to identify information flow problems in the current inspection process. A low-code platform is then used to integrate batch data, inspection specifications, staff capabilities, equipment constraints, and scheduling results. Based on the integrated data, a simulated annealing (SA) algorithm is applied to generate feasible inspection schedules under practical constraints.
To improve schedule adjustment under unexpected events, this study further compares two AI-assisted approaches. Model I uses a large language model (LLM) to directly generate adjusted schedules. Model II uses the LLM to parse event information and convert it into structured scheduling constraints, after which local simulated annealing is applied for rescheduling. The results show that Model II provides better feasibility, lower hallucination risk, and shorter computation time, indicating that generative AI is more suitable as an event interpretation and decision-support tool rather than as a direct replacement for scheduling algorithms.

Abdel-Aty, T. A., McFarlane, D., Brooks, S., Salter, L., Abubakar, A. S., Evans, S., . . . Mukherjee, A. (2024). The role of low-cost digital solutions in supporting industrial sustainability. Sustainability, 16(3), 1301.
An, Y., Chen, X., Gao, K., Zhang, L., Li, Y., & Zhao, Z. (2023). A hybrid multi-objective evolutionary algorithm for solving an adaptive flexible job-shop rescheduling problem with real-time order acceptance and condition-based preventive maintenance. Expert systems with applications, 212, 118711.
Bock, A. C., & Frank, U. (2021). Low-code platform. Business & Information Systems Engineering, 63(6), 733-740.
Borges, A. F., Laurindo, F. J., Spínola, M. M., Gonçalves, R. F., & Mattos, C. A. (2021). The strategic use of artificial intelligence in the digital era: Systematic literature review and future research directions. International journal of information management, 57, 102225.
Carrasco, J., García, S., Rueda, M. M., Das, S., & Herrera, F. (2020). Recent trends in the use of statistical tests for comparing swarm and evolutionary computing algorithms: Practical guidelines and a critical review. Swarm and Evolutionary Computation, 54, 100665.
Chen, T.-C. T., Chiu, M.-C., & Lin, Y.-C. (2025). Customizable GenAI system for assisting job scheduling in a flexible manufacturing system consisting of multiple factories. The International Journal of Advanced Manufacturing Technology, 1-14.
Das, J., Fisher, A. C., Hughey, L., O’Connor, T. F., Pai, V., Soto, C., & Wan, J. (2024). Considerations for Big Data management in pharmaceutical manufacturing. Current Opinion in Chemical Engineering, 46, 101051.
Defersha, F. M., Obimuyiwa, D., & Yimer, A. D. (2022). Mathematical model and simulated annealing algorithm for setup operator constrained flexible job shop scheduling problem. Computers & Industrial Engineering, 171, 108487.
Derrac, J., García, S., Molina, D., & Herrera, F. (2011). A practical tutorial on the use of nonparametric statistical tests as a methodology for comparing evolutionary and swarm intelligence algorithms. Swarm and Evolutionary Computation, 1(1), 3-18.
Ding, D. (2023). Transitioning to Microsoft Power Platform: An Excel User Guide to Building Integrated Cloud Applications in Power BI, Power Apps, and Power Automate. Springer.
Ellström, D., Holtström, J., Berg, E., & Josefsson, C. (2022). Dynamic capabilities for digital transformation. Journal of Strategy and Management, 15(2), 272-286.
Elqabli, Z., Kamach, O., Khatab, A., & Chater, Y. (2025). A novel dynamic scheduling model for application in multimode approach. Scientific Reports, 15(1), 27980.
Feuerriegel, S., Hartmann, J., Janiesch, C., & Zschech, P. (2024). Generative AI: S. Feuerriegel et al. Business & Information Systems Engineering, 66(1), 111-126.
Gui, Y., Tang, D., Zhu, H., Zhang, Y., & Zhang, Z. (2023). Dynamic scheduling for flexible job shop using a deep reinforcement learning approach. Computers & Industrial Engineering, 180, 109255.
Guo, D., & Mantravadi, S. (2025). The role of digital twins in lean supply chain management: review and research directions. International Journal of Production Research, 63(5), 1851-1872.
Gupta, S., Modgil, S., & Gunasekaran, A. (2020). Big data in lean six sigma: a review and further research directions. International Journal of Production Research, 58(3), 947-969.
Hajji, M. K., Hamlaoui, O., & Hadda, H. (2024). A simulated annealing metaheuristic approach to hybrid flow shop scheduling problem. Advances in Industrial and Manufacturing Engineering, 9, 100144.
Ham, A., Park, M.-J., & Kim, K. M. (2021). Energy‐Aware flexible job shop scheduling using mixed integer programming and constraint programming. Mathematical Problems in Engineering, 2021(1), 8035806.
Jeffery, S. (2025). The Low-Code AI Maturity Model: Leading Responsible AI Transformation with Microsoft Power Platform. Springer Nature.
Jeong, B., Han, J.-H., & Lee, J.-Y. (2021). Metaheuristics for a flow shop scheduling problem with urgent jobs and limited waiting times. Algorithms, 14(11), 323.
Liu, M., Lv, J., Du, S., Deng, Y., Shen, X., & Zhou, Y. (2024). Multi-resource constrained flexible job shop scheduling problem with fixture-pallet combinatorial optimisation. Computers & Industrial Engineering, 188, 109903.
Meng, X., Wang, N., Liu, J., & Liu, Q. (2022). An improved simulated annealing‐based decision model for the hybrid flow shop scheduling of aviation ordnance handling. Computational intelligence and neuroscience, 2022(1), 1843675.
Microsoft. (2026). 使用 Power Platform 進行應用程式現代化. Retrieved from https://learn.microsoft.com/zh-tw/power-platform/guidance/adoption/application-modernization
Miozza, M., Brunetta, F., & Appio, F. P. (2024). Digital transformation of the Pharmaceutical Industry: A future research agenda for management studies. Technological Forecasting and Social Change, 207, 123580.
Mirsanei, H., Zandieh, M., Moayed, M. J., & Khabbazi, M. R. (2011). A simulated annealing algorithm approach to hybrid flow shop scheduling with sequence-dependent setup times. Journal of Intelligent Manufacturing, 22(6), 965-978.
Novales, A., & Mancha, R. (2023). Fueling Digital Transformation with Citizen Developers and Low-Code Development. MIS Quarterly Executive, 22(3), 221-234.
Ohno, T. (2019). Toyota production system: beyond large-scale production: Productivity press.
Park, J.-S., Ng, H.-Y., Chua, T.-J., Ng, Y.-T., & Kim, J.-W. (2021). Unified genetic algorithm approach for solving flexible job-shop scheduling problem. Applied Sciences, 11(14), 6454.
Pedro, F., Veiga, F., & Mascarenhas-Melo, F. (2023). Impact of GAMP 5, data integrity and QbD on quality assurance in the pharmaceutical industry: How obvious is it? Drug Discovery Today, 28(11), 103759.
Peruzzini, M., Grandi, F., & Pellicciari, M. (2020). Exploring the potential of Operator 4.0 interface and monitoring. Computers & Industrial Engineering, 139, 105600.
Psarommatis, F., Prouvost, S., May, G., & Kiritsis, D. (2020). Product quality improvement policies in industry 4.0: characteristics, enabling factors, barriers, and evolution toward zero defect manufacturing. Frontiers in Computer Science, 2, 26.
Reddy, S. (2024). Generative AI in healthcare: an implementation science informed translational path on application, integration and governance. Implementation Science, 19(1), 27.
Roh, P., Kunz, A., & Wegener, K. (2019). Information stream mapping: Mapping, analysing and improving the efficiency of information streams in manufacturing value streams. CIRP Journal of Manufacturing Science and Technology, 25, 1-13.
Schumacher, S., Hall, R., Bildstein, A., & Bauernhansl, T. (2023). Lean Production Systems 4.0: systematic literature review and field study on the digital transformation of lean methods and tools. International Journal of Production Research, 61(24), 8751-8773.
Sufi, F. (2023). Algorithms in low-code-no-code for research applications: a practical review. Algorithms, 16(2), 108.
Tamssaouet, K., Dauzère-Pérès, S., Knopp, S., Bitar, A., & Yugma, C. (2022). Multiobjective optimization for complex flexible job-shop scheduling problems. European Journal of Operational Research, 296(1), 87-100.
Tan, J., Braubach, L., Jander, K., Xu, R., & Chen, K. (2020). A novel multi-agent scheduling mechanism for adaptation of production plans in case of supply chain disruptions. AI Communications, 33(1), 1-12.
Teng, G. (2025). An improved genetic algorithm for dual-resource constrained flexible job shop scheduling problem with tool-switching dependent setup time. Expert systems with applications, 281, 127496.
Tiwari, M. K., Bidanda, B., Geunes, J., Fernandes, K., & Dolgui, A. (2024). Supply chain digitisation and management. In (Vol. 62, pp. 2918-2926): Taylor & Francis.
Vărzaru, A. A., & Bocean, C. G. (2024). Digital transformation and innovation: The influence of digital technologies on turnover from innovation activities and types of innovation. Systems, 12(9), 359.
Vial, G. (2021). Understanding digital transformation: A review and a research agenda. Managing digital transformation, 13-66.
Wang, H., Chen, Y., Wang, L., Liu, Q., Yang, S., & Wang, C. (2023). Advancing herbal medicine: enhancing product quality and safety through robust quality control practices. Frontiers in pharmacology, 14, 1265178.
Yang, T., & Fujimoto, T. (2026). Jidoka just-in-time cycle–from genba kaizen to the introduction of artificial intelligence. International Journal of Production Research, 1-19.
Yunusoglu, P., & Topaloglu Yildiz, S. (2022). Constraint programming approach for multi-resource-constrained unrelated parallel machine scheduling problem with sequence-dependent setup times. International Journal of Production Research, 60(7), 2212-2229.