RISK BASED WORKER ALLOCATION AND LINE BALANCING A Thesis by Sathya Madhan Solaimuthu Pachaimuthu Bachelor of Engineering, Anna University, India, 2008 Submitted to the Department of Industrial and Manufacturing Engineering and the faculty of the Graduate School of Wichita State University in partial fulfillment of the requirements for the degree of Master of Science December 2010 i © Copyright 2010 by Sathya Madhan Solaimuthu Pachaimuthu All Rights Reserved ii RISK BASED WORKER ALLOCATION AND LINE BALANCING The following faculty members have examined the final copy of this thesis for form and content, and recommend that it be accepted in partial fulfillment of the requirement for the degree of Master of Science with a major in Industrial Engineering. ___________________________________ Krishna K. Krishnan, Committee Chair ___________________________________ Michael Jorgensen, Committee Member ___________________________________ Ramazan Asmatulu, Committee Member iii ABSTRACT In general, manufacturing systems could be classified into machine intensive manufacturing and labor intensive manufacturing. From the previous studies, we can infer that worker allocation plays an important role in determining efficiency of a labor intensive manufacturing system. Most of the research works in the previous literature is performed in a deterministic bed. But from the time study data that was obtained from a local aircraft company shows a high degree of variability in worker processing times. Thus this research presents a worker allocation approach which also considers the uncertainty in worker processing times into account. Risk based worker allocation approach is developed for three different scenarios. First scenario is the single task per station balanced production line scenario, where workers are allocated to processes by minimizing the overall risk of delay due to workers. In the second scenario, in addition to worker allocation by minimizing the overall risk, multiple workers are allocated to processes to make the flow of products uniform in a single task per station unbalanced production line. Prior to implementing the final approach, a method for line balancing when variability is involved is studied and compared to the ranked positional-weight method. The final scenario developed is a simultaneous approach to balance and allocate workers in a multiple task per station production line. Case studies were simulated using QUEST software and the result indicates that risk based allocation has increased throughput and efficiency compared to deterministic worker allocation. iv TABLE OF CONTENTS Chapter Page 1 INTRODUCTION ................................................................................................................ 1 1.1 Problem Background ............................................................................................... 1 1.2 Worker Allocation Problem .................................................................................... 2 1.3 Worker Profile ......................................................................................................... 2 1.3.1 Sole Profile Allocation ................................................................................ 3 1.3.2 Multi Profile Allocation .............................................................................. 4 1.4 Research Purpose .................................................................................................... 6 1.5 Research Objective .................................................................................................. 8 2 LITERATURE REVIEW ................................................................................................... 10 2.1 Introduction ........................................................................................................... 10 2.2 Production Lines ................................................................................................... 10 2.3 Worker Allocation ................................................................................................. 11 2.4 Sole Profile Allocation .......................................................................................... 11 2.4.1 Formulations in Sole Profile allocation ..................................................... 11 2.5 Multi Profile Allocation ........................................................................................ 14 2.5.1 Formulations in Multi Profile allocation ................................................... 14 2.6 Research Motivation ............................................................................................. 18 3 RISK BASED WORKER ALLOCATION METHOD ..................................................... 19 3.1 Introduction ........................................................................................................... 19 3.2 Notations ............................................................................................................... 19 3.3 Worker Allocation Procedure................................................................................ 20 3.3.1 Worker Allocation in Single Task per Station Balanced Line .................. 21 3.3.1.1 Worker Allocation without Considering Uncertainty ................... 21 3.3.1.1.1 Model Input Requirements ............................................. 22 3.3.1.1.2 Deterministic Mathematical Model ................................ 22 3.3.1.1.3 Case Study–3-Process-6-Worker (Deterministic) .......... 24 3.3.1.2 Worker Allocation Considering Uncertainty ................................ 27 3.3.1.2.1 Worker Allocation Risk Assessment Procedure ............ 27 v TABLE OF CONTENTS (Continued) Chapter ...................................................................................................................................... Page 3.3.1.2.2 Model Input Requirements ............................................. 28 3.3.1.2.3 Risk Based Mathematical Model ................................... 28 3.3.1.2.4 Case Study – 3-Process-6-Worker (Risk Based)............ 30 3.3.1.2.5 Case Study – 4-Process-8-Worker ................................. 32 3.3.1.3 Validation ...................................................................................... 35 3.3.2 Worker Allocation in Single Task per Station Un-Balanced Line ............ 38 3.3.2.1 Model Input Requirements ............................................................ 38 3.3.2.2 Mathematical Model for Unbalanced Line ................................... 39 3.3.2.3 Case Study - 5-Process-10-Worker ............................................... 41 3.3.2.4 Case Study - 5-Process-15-Worker ............................................... 47 3.4 Conclusion ............................................................................................................. 50 4 RISK BASED LINE BALANCING .................................................................................. 51 4.1 Introduction ........................................................................................................... 51 4.2 Notations ............................................................................................................... 51 4.3 Assumptions .......................................................................................................... 52 4.4 Model Input Requirements .................................................................................... 52 4.5 Mathematical Model ............................................................................................. 52 4.6 Case Study – 9-Task-3-Station .............................................................................. 54 4.7 Validation .............................................................................................................. 57 4.8 Optimum Number of Station Determination ......................................................... 58 4.9 Case Study – 9 Task .............................................................................................. 60 4.10 Conclusion ............................................................................................................. 62 5 SIMULTANEOUS RISK BASED LINE BALANCING AND WORKER ALLOCATION ....................................................................................................................................................... 63 5.1 Introduction ........................................................................................................... 63 5.2 Notations ............................................................................................................... 63 5.3 Assumptions .......................................................................................................... 64 5.4 Model Input Requirements .................................................................................... 64 5.5 Mathematical Model ............................................................................................. 65 vi TABLE OF CONTENTS (Continued) Chapter ...................................................................................................................................... Page 5.6 Case Study – 9 Task – 6 Worker ........................................................................... 69 5.7 Conclusion ............................................................................................................. 78 6 CONCLUSION AND FUTURE RESEARCH .................................................................. 80 6.1 Conclusion ............................................................................................................. 80 6.2 Intellectual Merit ................................................................................................... 82 6.3 Future Research ..................................................................................................... 83 REFERENCES .............................................................................................................................. 85 APPENDICES ............................................................................................................................... 88 A: LINGO model for worker allocation without considering uncertainty -3-process-6-worker .. 89 B: LINGO model for worker allocation considering uncertainty -3-process-6-worker................ 91 C: LINGO model for worker allocation considering uncertainty -4-process-8-worker................ 93 D: LINGO model for worker allocation unbalanced line -5-process-10-worker .......................... 95 E: LINGO model for worker allocation unbalanced line -5-process-15-worker .......................... 97 F: LINGO model for risk based line balancing-9-task-3-station ................................................ 100 G: LINGO model for risk based line balancing-9-task-4-station ................................................ 101 H: LINGO model for risk based line balancing-9-task-5-station ................................................ 102 I: LINGO model for SRLW-9-task-6-worker-3-station .............................................................. 103 J: LINGO model for SRLW-9-task-6-worker-4-station.............................................................. 105 K: LINGO model for SRLW-9-task-6-worker-5-station ............................................................ 107 vii LIST OF TABLES Table Page 1.1 Formulations - Sole Profile Allocation 3 1.2 Formulations - Multi Profile Allocation 5 1.3 Sample Task Processing Time Data 7 3.1 Processing Time Chart for 3-Process-6-Worker Case Study (Deterministic) 24 3.2 Quality Level Chart for 3-Process-6-Worker Case Study (Deterministic) 25 3.3 Worker Cost Chart for 3-Process-6-Worker Case Study (Deterministic) 25 3.4 Results - 3-Process-6-Worker Case Study (Deterministic) 26 3.5 Processing Time Chart for 3-Process-6-Worker Case Study (Risk) 30 3.6 Quality Level Chart for 3-Process-6-Worker Case Study (Risk) 30 3.7 Results - 3 Process 6 Worker Case study (Risk) 32 3.8 Processing Time Chart for 4-Process-8-Worker Case Study 33 3.9 Quality Level Chart for 4-Process-8-Worker Case Study 33 3.10 Worker Cost ($) Chart for 4-Process-8-Worker Case Study 34 3.11 Results – 4-Process-8-Worker Case Study 35 3.12 Cost Comparison Between Deterministic and Risk Based Method 37 3.13 VA/NVA Comparison Between Deterministic and Risk Based Method 37 3.14 Processing Time (min) Chart for 5-Process-10-Worker Case Study 41 3.15 Quality Level Chart for 5-Process-10-Worker Case Study 42 3.16 Worker Cost ($) Chart for 5-Process-10-Worker Case Study 43 3.17 Results – 5-Process-10-Worker Case Study 46 3.18 Processing Time (min) Chart for 5-Process-15-Worker Case Study 47 viii LIST OF TABLES (continued) Table Page 3.19 Quality Level Chart for 5-Process-15-Worker Case Study 48 3.20 Worker Cost ($) Chart for 5-Process-15-Worker Case Study 48 3.21 Results – 5-Process-10-Worker Case study 49 4.1 Expected Processing Time for 9-Task-3-Station Case Study 54 4.2 Processing Time (min) Chart for 9-Task–3-Station Case Study 55 4.3 Results –9-Task–3-Station-Case Study-Risk Based Allocation 56 4.4 Results –9-Task–3-Station-Case Study-Deterministic Allocation 57 4.5 Comparison Between Deterministic and Risk Based Method 57 4.6 Results –9-Task–3-Station-Case Study 60 4.7 Results –9-Task–4-Station-Case Study 61 5.1 Expected Processing Time/ Quality Level for 9-Task–6-Worker Case Study 70 5.2 Processing Time (min) Chart for 3-Station-9-Task–6-Worker Case Study 70 5.3 Quality Level Chart for 3-Station-9-Task–6-Worker Case Study 71 5.4 Worker Cost ($) Chart for 3-Station-9-Task–6-Worker Case Study 71 5.5 Results – 3-Station-9-Task–6-Worker Case Study 74 5.6 Results – 4-Station-9-Task–6-Worker Case Study 76 5.7 Results – 5-Station-9-Task–6-Worker Case Study 78 ix LIST OF FIGURES Figure Page 2.1 Performance of algorithms (Askin and Huang, 2001) 15 3.1 Case Study - 3-Process-6-Workers 24 3.2 Case Study - 4-Process-8-Workers 32 3.3 VA/NVA Comparison Chart 38 3.4 Case Study - 5-Process-10-Workers 41 3.5 Case Study - 5 Process and 15 Worker 47 4.1 Cycle Time Comparison Chart 58 4.2 Flowchart for optimal number of station determination - RLB 59 5.1 Flowchart for optimal number of station determination – SRLW 69 x