چشم‌انداز مدیریت صنعتی

چشم‌انداز مدیریت صنعتی

ارائه مدل پیش‌بینی ریسک‌های بحرانی شبکه انتقال گاز با استفاده از الگوریتم‌های داده‌کاوی

نوع مقاله : مقاله پژوهشی

نویسندگان
محمد صادق بهروز 1
محمدعلی افشار کاظمی 2
عادل آذر 3
عزت‌اله اصغری‌زاده 4
1 دانشجوی دکتری مدیریت صنعتی، ‌گروه مدیریت صنعتی، دانشکده مدیریت و اقتصاد، واحد علوم و تحقیقات، دانشگاه آزاد اسلامی، تهران، ایران.
2 دانشیار، گروه مدیریت صنعتی، دانشکده مدیریت و اقتصاد، واحد علوم و تحقیقات، دانشگاه آزاد اسلامی، تهران، ایران.
3 استاد، گروه مدیریت صنعتی، دانشکده مدیریت و اقتصاد، دانشگاه تربیت مدرس، تهران، ایران.
4 دانشیار، گروه مدیریت، دانشکده مدیریت، دانشگاه تهران، تهران، ایران.
10.48308/jimp.13.1.281
چکیده
باتوجه به نقش مهم رویکردهای پیش­بینانه در کاهش هزینه‌­های نگهداری تعمیرات، هدف از انجام پژوهش، ارائه مدل پیش­بینی ریسک­‌های بحرانی و اولویت­‌دار بر پایه الگوریتم­‌های داده­‌کاوی است. روش داده­‌کاوی پژوهش بر اساس روش CRISP طرح‌­ریزی شده است. مدل‌سازی داده­‌ها بر پایه داده‌­کاوی «توصیفی» و«پیش­بینی» و استفاده از الگوریتم­‌های خوشه‌­بندی و طبقه­‌بندی است. شاخص سیلوئیت مبنای خوشه‌­بندی در نظر گرفته شده و از الگوریتم‌­های Two Step، Kohnen و K-Means استفاده شده است. بهترین مقدار، مبتنی بر الگوریتم K-Means برابر 6446/0 با تعداد خوشه 5 بود و ویژگی­‌های اصلی برای انجام طبقه­‌بندی و پیش­بینی ریسک­‌ها تعیین شد. الگوریتم­‌های شبکه عصبی، درخت C.5، نزدیک‌ترین همسایگی و بردار پشتیبان برای طبقه‌­بندی استفاده شده است. در این پژوهش، الگوریتم­ ترکیبی پیش­بینی به‌صورت تکاملی به‌کارگیری شده و در هر مرحله، هدف تقویت میزان صحت و اعتبار مدل طبقه­‌بندی و افزایش یادگیری داده­‌ها است. نتایج پژوهش، یادگیری در56/97 درصد از داده­‌های موردتوافق را نشان داده و میزان صحت و اعتبار مدل ترکیبی برای طبقه­‌بندی داده‌­ها، 86/92 درصد برآورد شده است. بر اساس نتایج، 13 ریسک، بحرانی تشخیص داده شده‌­اند که در این میان «انتشار گازهای آلاینده و مواد شیمیایی» و «عدم‌­آموزش و توجیه­‌نبودن پیمانکاران نسبت به موقعیت شبکه» به‌­ترتیب بیشترین و کمترین اولویت را دارد.
کلیدواژه‌ها
ارزیابی ریسک
نگهداری و تعمیرات
مدل‌سازی
داده‌کاوی
شبکه انتقال گاز
موضوعات

عنوان مقاله English

Prediction Model of the Gas Pipeline Critical Risk Using Data Mining Algorithms

نویسندگان English

Mohammad Sadegh Behrouz 1
Mohammad Ali AfsharKazemi 2
Adel Azar 3
Ezatolah Asgharizadeh 4
1 Ph.D. Student of Industrial Management, Faculty of Management and Economics, Branch of Science and Research, Islamic Azad University, Tehran, Iran.
2 Associate Professor, Faculty of Management and Economics, Branch of Science and Research, Islamic Azad University, Tehran, Iran.
3 Professor, Management and Accounting Faculty, Tarbiat Modares University, Tehran, Iran.
4 Associate professor, Faculty of management, Tehran University, Tehran, Iran.
چکیده English

Predictive approaches play an important role in detecting events, controlling risks and reducing maintenance and repair costs. The purpose is to provide a model for predicting critical and prioritized risks based on data mining algorithms. Data mining method was planned based on the CRISP methodology. Data modeling has been done in two parts: "descriptive" and "predictive" data mining and the use of "clustering" and "classification" algorithms."Sillhouette index" is considered for clustering and the K-Means, Kohnen, Two Step algorithm is used; the best value is based on the K-Means algorithm. Silhouette is equal to 0.6446 with the number of clusters 5. Next, Neural Network Algorithms, C.5 tree, Nearest Neighbor and Support Vector have been used for classification. These techniques recognizing data classification patterns and their integration increases the amount of data learning. The results showed learning in 97.56% of the agreed data and the accuracy and validity of the combined model for data classification was estimated at 92.86%. Based on the results, 13 critical risks have been identified; "release of polluting gases and chemicals" and "lack of training and justification of contractors regarding the pipeline" have the highest and lowest priority, respectively.

کلیدواژه‌ها English

Risk Assessment
Maintenance
Modeling
Data Mining
Gas Pipeline
  1. Abdel-aziz, I. H., & Helal, M. (2012). Application of FMEA- FTA in Reliability- Centered Maintenance Planning. 15th International conferance on Applied Mechanics and Mechanicals Engineering.
  2. Alborzi, M. (2013). Neural Networks. Publications of Sharif University of Technology, 5Ed, Tehran, Iran (In Persian).
  3. Alikhanzadeh, A. (2012). Data Mining. Oloom Computer, 3Ed, Babol, Iran (In Persian).
  4. Alizadeh, S., & Malek Mohammadi, S. (2013). Data mining and knowledge discovery step by step with Clementine software. Publications of Khwaja Nasiruddin Toosi University of Technology, 3Ed, Tehran, Iran (In Persian).
  5. Ameri, H., Alizadeh, S., & Barzegari, A. (2012). Extracting knowledge from the data of diabetic patients using decision tree method C5. Health Management, 16(53), 58-72. (In Persian)
  6. Ardeshir, A., Amiri, M., & Mohajer, M. (2012). Assessing safety risks in mass production projects using the combination of fuzzy AHP-DEA and fuzzy FTA methods, FMEA. Bimonthly work health of Iran, 10(6), 78-91. (In Persian)
  7. Bashiri Nasab, M., Gholamreza, A., & Farzaneh, S. (1389). Safety Management. Fanavaran Publishing, 1Ed, Tehran, Iran (In Persian).
  8. Chang, H. H., & Tsay, S. F. (2009). Integrating of SOM and K-mean in data mining clustering: An empirical study of CRM and profitability evaluation.
  9. Chang, W., Meng, T., & Lim, C. (2015). Clustering and visualization of failure modes using an evolving tree. Expert Systems with Applications, 42, 7235–7244.
  10. Chi, C. F., Sigmund, D., & Octavianus, M. (2020). Classification Scheme for Root Cause and Failure Modes and Effects Analysis (FMEA) of Passenger Vehicle Recalls. Reliability Engineering and System Safety, 200, 906-929.
  11. Dermohammadi, S., Alizadeh, S., Asghari, M., & Shami, M. (2013). Providing a predictive model for diagnosing infertility factors; Using data mining algorithms. Health Management, 17(57), 46-57. (In Persian)
  12. Fahmi Hassan, A., Mughari, M., & Obadai, O. (1397). Prediction of blood donation using data mining based on decision tree algorithms, KNN, SVM, MLP. Journal of engineering management and soft computing, 4(1), 77-97. (In Persian)
  13. Ghasemi, Sh., Yavari, K., Mahmoud-Vand, R., & Sahabi, B. (2013). Comparison of two different viewpoints in the application of FMEA method for risk assessment: a case study of Iranian gas refinery. Quarterly journal of energy economy studies, 10(42), 159-135 (In Persian).
  14. Ghazanfari, M., Alizadeh, S., & Timurpur, B. (2015). Data mining and knowledge discovery. University of Science and Technology Publication. 5Ed, Tehran, Iran (In Persian).
  15. Ghodoosi, M., Mirsaeedi, F., & Hasani, A. (2020). Presentation of Risks Analysis Model in Urban Projects Based on Data Mining Technique with Case Study. The Journal of Industrial Management Perspective, 10, 137-159. (In Persian)
  16. Habibi, E., & Alizadeh, M. (1386). Functional safety and performance indicators in industry. Fanavaran Publishing, 2Ed, Tehran, Iran. (In Persian)
  17. Hamta, N., Ghobadi, Sh., & Boalhosni, P. (2017). Improving reliability in continuous manufacturing industries by applying data mining, FMEA and FTA approaches. The first business management and optimization systems conference. Babol. Iran (In Persian).
  18. Hosseini al-Madwari, M., Moghdasi, M. & Shafiizadeh Bafghi, M. (1390). Risk assessment by FMEA method and comparison of RPN before and after corrective measures in Bafq steel direct revitalization project. The 7th national conference on occupational health and safety. (In Persian)
  19. Jafari Naimi, M. (1390). Clustering of digestive functional disorders. Master's thesis, Technical and Engineering Faculty, Isfahan University. (In Persian)
  20. Mohaghar, A., & Khorasani, A. (2020). Designing the Model of Assessing the Risk management of Automaker Companies in Iran: Grounded Theory. The Journal of Industrial Management Perspective, 10, 137-159 (In Persian).
  21. Moniri, M., Alem-Tabriz, A., & Ayough, A. (2022). Upstream Oil Process Plants Turnaround Projects Risk Evaluation Using a Hybrid Fuzzy MADM Method. The Journal of Industrial Management Perspective, 12, 135-173 (In Persian).
  22. Morovati Sharif Abadi, A., Zanjirchi, M., & Abbasabadi, O. (2022). Processes Management of Maintenance using PCF and Data Mining. The Journal of Industrial Management Perspective, 12, 175-198. (In Persian)
  23. Ramezani Beshli, P. (2011). Cluster analysis and analysis of foreshocks for earthquake prediction using data-mining techniques. Master's Thesis, Faculty of Earth Sciences, Shahrood University of Technology (In Persian).
  24. Ravanbakhsh, S. (2017). Improving the efficiency of strategic equipment with the method of maintenance, damage analysis and simulation. Journal of maritime transportation industry, 4(3), 11-20. (In Persian).
  25. Sabet Mutlaq, M., Ayazi, A., & Hosseini Dashiri, Jalaluddin. (2016). Presenting a hybrid approach to evaluate and rank failure modes using modified FMEA and fuzzy hierarchical analysis process. Case study: A company producing gears and industrial gearboxes active in Qom. Scientific and Promotional Quarterly of Standard and Quality Management, 7(3), 19-30. (In Persian).
  26. Sheikhinejad, Z., Naderi Dehkordi, M., & Rastgari, H. (2013). A review of privacy-preserving methods in transactional database security. The second computer science and education conference. (In Persian)
  27. Souza, R., & Alvares, A. (2007). FMEA and FTA Analysis for Application of the Reliability Centered Maintenance Methodology. 19th International congress of Mechanical Engineering. Brasilia.
  28. Steenwinckel, B., De Paepe, D., Heyvaert, P., & Moens, P. (2021). FLAGS: A methodology for adaptive anomaly detection and root cause analysis on sensor data streams by fusing expert knowledge withmachine learning. Future Generation Computer Systems, 116, 30–48.
  29. Yang, C., Zou, Y., Lai, P., & Jiang, N. (2015). Data mining-based methods for fault isolation with validated FMEA model ranking. Springer.