Machine learning for weekly work plan generation and enhanced constraint predication in last planner system (Softcopy is also available)
Gandhi, Jaivik (PCM22145)
Machine learning for weekly work plan generation and enhanced constraint predication in last planner system (Softcopy is also available) - 2024 - xviii,98p.
Abstract I
Undertaking Iii
Certificate V
Acknowledgements Vii
Abbreviations Ix
Table Of Contents Xi
List Of Figures Xiii
List Of Tables Xv
List Of Appendix Xvii
Chapter-1: Introduction 1
1.1 Need For The Study 1
Chapter-2: Literature Review 3
2.1.1 Lean Construction 3
2.1.2 Last Planner System 4
2.1.3 Machine Learning And Lean Construction 5
2.1.4 Application Of Ml In Construction Industry 6
2.1.5 Analysis Of Algorithms In Machine Learning 8
2.1.6 Synthetic Data 9
2.2 Previous Study 11
2.3 Aims And Objectives 11
2.4 Research Methodology For Research 12
Chapter-3: Research Methodology 15
3.1 Analysis Of The Research Methodology 15
3.1.1 Literature Review 16
3.1.2 Current Practices In Lps, Machine Learning And Synthetic Data Generation 16
3.1.3 Data Collection 17
3.1.4 Feature Selection 17
3.1.5 Label Encoding 17
3.1.6 Model For Synthetic Data Generation 17
3.1.7 Developing The Prediction Model 18
Chapter-4: Data Collection 19
4.1 Data Of Super-Speciality Healthcare Project 19
4.2 Constraint Log 19
4.3 Constraints Analysis 20
4.4 Data Collection Summary 21
Chapter-5: Data Analysis 23
5.1 Synthetic Data Generation 23
5.2 Methods Used To Generate The Data 23
5.2.1 Gretel.Ai 23
5.2.2 Mostly.Ai 25
5.2.3 Github (Synthetic Data Vault) 26
5.3 Synthetic Data Generation For Single Activity 26
5.3.1 Development Of The Code For One Activity 27
5.3.2 Final Code For All Activities 30
5.4 Ml Model Preparation For Prediction 33
5.4.1 Importing Various Library 33
5.4.2 Files In Google Drive 34
5.4.3 Training The Model 35
5.5 Ml Model 1(Sgd) 37
5.5.1 Output Of The Model 38
5.5.2 Hyperparameter Tuning 38
5.6 Testing The Model 39
5.6.1 Final Output 40
5.7 Ml Model 2(Multiout Regressor) 45
5.7.1 Output Of The Model 47
Chapter-6: Conclusion 51
6.1 Future Scope 53
References 55
Appendix 61
CEM TH-0457 / GAN
Machine learning for weekly work plan generation and enhanced constraint predication in last planner system (Softcopy is also available) - 2024 - xviii,98p.
Abstract I
Undertaking Iii
Certificate V
Acknowledgements Vii
Abbreviations Ix
Table Of Contents Xi
List Of Figures Xiii
List Of Tables Xv
List Of Appendix Xvii
Chapter-1: Introduction 1
1.1 Need For The Study 1
Chapter-2: Literature Review 3
2.1.1 Lean Construction 3
2.1.2 Last Planner System 4
2.1.3 Machine Learning And Lean Construction 5
2.1.4 Application Of Ml In Construction Industry 6
2.1.5 Analysis Of Algorithms In Machine Learning 8
2.1.6 Synthetic Data 9
2.2 Previous Study 11
2.3 Aims And Objectives 11
2.4 Research Methodology For Research 12
Chapter-3: Research Methodology 15
3.1 Analysis Of The Research Methodology 15
3.1.1 Literature Review 16
3.1.2 Current Practices In Lps, Machine Learning And Synthetic Data Generation 16
3.1.3 Data Collection 17
3.1.4 Feature Selection 17
3.1.5 Label Encoding 17
3.1.6 Model For Synthetic Data Generation 17
3.1.7 Developing The Prediction Model 18
Chapter-4: Data Collection 19
4.1 Data Of Super-Speciality Healthcare Project 19
4.2 Constraint Log 19
4.3 Constraints Analysis 20
4.4 Data Collection Summary 21
Chapter-5: Data Analysis 23
5.1 Synthetic Data Generation 23
5.2 Methods Used To Generate The Data 23
5.2.1 Gretel.Ai 23
5.2.2 Mostly.Ai 25
5.2.3 Github (Synthetic Data Vault) 26
5.3 Synthetic Data Generation For Single Activity 26
5.3.1 Development Of The Code For One Activity 27
5.3.2 Final Code For All Activities 30
5.4 Ml Model Preparation For Prediction 33
5.4.1 Importing Various Library 33
5.4.2 Files In Google Drive 34
5.4.3 Training The Model 35
5.5 Ml Model 1(Sgd) 37
5.5.1 Output Of The Model 38
5.5.2 Hyperparameter Tuning 38
5.6 Testing The Model 39
5.6.1 Final Output 40
5.7 Ml Model 2(Multiout Regressor) 45
5.7.1 Output Of The Model 47
Chapter-6: Conclusion 51
6.1 Future Scope 53
References 55
Appendix 61
CEM TH-0457 / GAN