Scholarlink

<p>Modern Manufacturing Execution Systems (MES) produce vast amounts of event data, yet translating this data into meaningful, actionable insights remains a significant challenge. This study addresses that gap by applying Process Mining techniques to both real-world and simulated manufacturing environments, aiming to improve process transparency, operational efficiency, and data-driven decision-making. Using methodologies such as Alpha Miner, Heuristic Miner, and Genetic Miner, the research uncovers inefficiencies, ensures process compliance, and enhances performance across various industrial domains. A key case study involving Samsung Electro-Mechanics revealed underlying process inefficiencies and machine workload imbalances, despite initially high model fitness scores. To further validate the findings, a custom simulation built using GoLang was employed, successfully confirming the accuracy of the Alpha Miner approach. This dual application to both actual and simulated data underscores the potential of Process Mining in optimizing manufacturing workflows through precise event log analysis. The study’s novelty lies in its integration of a bespoke GoLang-based event log generator and its demonstration of Process Mining’s adaptability across multiple domains. Theoretical contributions include the evaluation of process model accuracy and the establishment of a validated simulation framework. From a practical perspective, the study supports enhanced manufacturing efficiency, informed decision-making, and the development of risk-free training environments. By transforming raw MES data into structured, interpretable insights, this research offers valuable implications for both industry practitioners and academic researchers.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Unveiling the dynamics of business processes: the role of process mining in enhancing efficiency, ensuring compliance, and driving continuous improvement

  • 期刊论文

  • 发表日期：

    2025年7月29日

查看全文
  • Habibullah Faisal,
  • Sk Ayub Al Wahid,
  • Sufi Sudruddin Chowdhury,
  • Md Sohel Rana,
  • Tisha Farhana

摘要

Modern Manufacturing Execution Systems (MES) produce vast amounts of event data, yet translating this data into meaningful, actionable insights remains a significant challenge. This study addresses that gap by applying Process Mining techniques to both real-world and simulated manufacturing environments, aiming to improve process transparency, operational efficiency, and data-driven decision-making. Using methodologies such as Alpha Miner, Heuristic Miner, and Genetic Miner, the research uncovers inefficiencies, ensures process compliance, and enhances performance across various industrial domains. A key case study involving Samsung Electro-Mechanics revealed underlying process inefficiencies and machine workload imbalances, despite initially high model fitness scores. To further validate the findings, a custom simulation built using GoLang was employed, successfully confirming the accuracy of the Alpha Miner approach. This dual application to both actual and simulated data underscores the potential of Process Mining in optimizing manufacturing workflows through precise event log analysis. The study’s novelty lies in its integration of a bespoke GoLang-based event log generator and its demonstration of Process Mining’s adaptability across multiple domains. Theoretical contributions include the evaluation of process model accuracy and the establishment of a validated simulation framework. From a practical perspective, the study supports enhanced manufacturing efficiency, informed decision-making, and the development of risk-free training environments. By transforming raw MES data into structured, interpretable insights, this research offers valuable implications for both industry practitioners and academic researchers.