VIETNAM NATIONAL UNIVERSITY HO CHI MINH CITY HO CHI MINH CITY UNIVERSITY OF TECHNOLOGY TRUONG CONG HUY ANALYZE THE IMPACT OF FACTORS AFFECTING REWORK COSTS IN CONSTRUCTION PROJECTS USING ANN Major: Construction Management Major code: 8580302 MASTER’S THESIS HO CHI MINH CITY, June 2024 THIS RESEARCH IS COMPLETED AT UNIVERSITY OF TECHNOLOGY – VNU – HCM CITY Instructor: Assoc. Tran Duc Hoc The examiner 1: Assoc. Do Tien Sy The examiner 2: Dr. Nguyen Thanh Viet Master’s thesis is defended at HCM city University of Technology, VNU- HCM City on June 20th, 2024.
The board of the Master’s Thesis Defense Council includes: 1. Le Hoai Long 2. Huynh Nhat Minh 3. Do Tien Sy 4.
Nguyen Thanh Viet 5. Council member: Assoc. Quek Tiang Beng Verification of the Chairman of the Master’s Thesis Defense Council and the Dean of faculty of Civil Engineering after the thesis being corrected (If any). CHAIRMAN OF THESIS COMMITTE HEAD OF FACULTY OF CIVIL ENGINEERING Dr Le Hoai Long i VIETNAM NATIONAL UNIVERSITY HO SOCIALIST REPUBLIC OF VIETNAM CHI MINH CITY Independence – Freedom - Happiness HO CHI MINH CITY UNIVERSITY OF TECHNOLOGY THE TASK SHEET OF MASTER’S THESIS Full name: TRUONG CONG HUY Student code: 2270240 Date of birth: 06th November 1996 Place of birth: HCMC Major: Construction Management Major code : 8580302 I.
THESIS TOPIC TÊN ĐỀ TÀI: Analyze the impact of factors affecting rework costs in construction projects using ANN (Tạm dịch: Phân tích tác động của các yếu tố ảnh hưởng đến chi phí làm lại trong các dự án xây dựng sử dụng ANN) II. TASKS AND CONTENTS: Rework cost factors and rework cost estimation by ANN Model III. TASKS STARTING DATE : 15th January 2024 IV. TASKS ENDING DATE: 20th May 2024 V.
Tran Duc Hoc HCM City, date. INSTRUCTOR HEAD OF DEPARTMENT (Full name and signature) (Full name and signature) Assoc. Dr Tran Duc Hoc DEAN OF FACULTY OF CIVIL ENGINEERING (Full name and signature) ii ACKNOWLEDGEMENT During a few month of attempting for studying and researching, I already finished my graduation project with title “Analyze the impact of factors affecting rework costs in construction projects using ANN”. This graduation were completed not only by author effort but also all surrounding people support.
First of all, I would like to give my deepest and warmest thanks to Assoc. Tran Duc Hoc, my instructor. Your brilliant comments, suggestions and connections have been extremely helpful for me, not only on this journey of undertaking my master’s thesis research, but also in my future career and life. Secondly, I sincerely thank to lecturers of Construction Management Department, HCMC University of Technology enthusiastically taught and inspire valuable knowledge to me during my 2 years of study at university.
Furthermore, I give my big thank to my classmates of International Master Program – Ho Chi Minh City University of Technology, who have made this program a huge success and continuously supported me during the course. I would like to send my special thanks to the experts, participants on survey, my colleague at GS E&C Company and on Thu Thiem Zeit River project such as project management unit, IVY contractor, ICIC consultant, engineer from contractors during my working time on bidding department who have facilitated their precious time and efforts to support my survey and shared with me their best advice. Last but not least, I am forever grateful to my family members and friends, who have always been by my side, caring, understanding and giving me valuable encouragement during difficult times. Best regards Ho Chi Minh City, , 2024 TRUONG CONG HUY iii ABSTRACT "Rework" has long been recognized as a significant challenge in construction projects.
While several studies have identified key factors that contribute to rework costs, the aim of this graduation project is to pinpoint the major contributors to rework costs through a survey-based questionnaire method. Additionally, the construction sector's impact on the environment and natural resource consumption is substantial, with significant energy use during construction and operation. Rework exacerbates these issues by generating additional waste and increasing resource use. This research explores the relationship between factors like design changes, communication issues, and other influences from various construction stakeholders including investors, consultants, contractors, and environmental considerations and project efficiency as measured by rework costs.
Despite the clear link between these factors and rework, there's currently no established tool to quantify these relationships. This study addresses that gap by developing an artificial neural network (ANN) model using RapidMiner Studio to evaluate and predict how communication-related factors impact rework costs in construction projects. The model serves as an integrated decision support tool for project management, highlighting the differences between actual and predicted rework costs in completed projects in Vietnam from 2013 to 2023. Ultimately, this research is intended to help construction stakeholders make better-informed decisions regarding rework, providing a reference point for addressing rework challenges in the construction industry.
During this graduation project, total of 43 factors were divided into 07 component group including others and Vietnam specific characteristics, Design Issue, Contractor/Subcontractor, Investor, Labour and Equipment, Supervisory, Environment and external impact. In addition, collected data of 86 project were used for setting ANN Data on Rapidminer Studio software. ANN design structure including 3 layers: input layer (14 neurons), Hidden layer (4 neurons) and output layer (1 neuron) with Sigmoid function. As a consequence, ANN prediction model for rework cost with the highest correlation coefficient value so ANN Model is more reliable than LR, CART, SVM model.
Keyword: Rework cost, Impact of factors, Artificial Neural Network, Machine Learning, Artificial Intelligence. iv TÓM TẮT LUẬN VĂN "Công tác làm lại" từ lâu đã được coi là một thách thức đáng kể trong các dự án xây dựng. Trong khi một số nghiên cứu đã xác định các yếu tố chính góp phần vào chi phí làm lại, mục tiêu của đồ án tốt nghiệp này là xác định chính xác những yếu tố góp phần chính vào chi phí làm lại thông qua phương pháp bảng câu hỏi dựa trên khảo sát. Ngoài ra, tác động của ngành xây dựng đối với môi trường và mức tiêu thụ tài nguyên thiên nhiên là rất lớn, với việc sử dụng năng lượng đáng kể trong quá trình xây dựng và vận hành.
Các công tác làm lại gây trầm trọng thêm các vấn đề này bằng cách tạo ra thêm chất thải và tăng mức sử dụng tài nguyên. Luận văn này khám phá mối quan hệ giữa các yếu tố như thay đổi thiết kế, vấn đề giao tiếp và các tác động khác từ nhiều bên liên quan trong lĩnh vực xây dựng bao gồm nhà đầu tư, tư vấn, nhà thầu và các cân nhắc về môi trường cũng như hiệu quả của dự án được đo bằng chi phí làm lại. Mặc dù có mối liên hệ rõ ràng giữa các yếu tố này và việc làm lại, nhưng hiện tại vẫn chưa có công cụ nào được thiết lập để định lượng các mối quan hệ này. Nghiên cứu này giải quyết sự thiếu hụt đó bằng cách phát triển mô hình mạng nơ-ron nhân tạo (ANN) sử dụng RapidMiner Studio để đánh giá và dự đoán cách các yếu tố liên quan đến giao tiếp tác động đến chi phí làm lại trong các dự án xây dựng.
Mô hình ANN đóng vai trò là công cụ hỗ trợ quyết định tích hợp cho quản lý dự án, làm nổi bật sự khác biệt giữa chi phí làm lại thực tế và dự kiến trong các dự án đã hoàn thành tại Việt Nam từ năm 2013 đến năm 2023. Cuối cùng, nghiên cứu này nhằm mục đích giúp các bên liên quan trong ngành xây dựng đưa ra quyết định sáng suốt hơn về công tác làm lại, cung cấp điểm tham chiếu để giải quyết các thách thức về công tác làm lại trong ngành xây dựng. Trong đồ án tốt nghiệp này, tổng cộng 43 yếu tố được chia thành 07 nhóm chính bao gồm các yếu tố khác và đặc trưng riêng của Việt Nam, thiết kế, nhà thầu/nhà thầu phụ, chủ đầu tư, nhân công và thiết bị, giám sát, môi trường và các tác động bên ngoài. Ngoài ra, dữ liệu thu thập được của 86 dự án đã được sử dụng để thiết lập dữ liệu ANN trên phần mềm Rapidminer Studio.
Cấu trúc thiết kế ANN bao gồm 3 lớp: lớp đầu vào (14 nơ-ron), Lớp ẩn (4 nơ-ron) và lớp đầu ra (1 nơ-ron) với hàm Sigmoid. Do đó, mô hình ANN dự đoán chi phí làm lại có giá trị hệ số tương quan cao nhất nên Mô hình ANN đáng tin cậy hơn mô hình LR, CART, SVM. v THE COMMITMENT OF AUTHOR The undersigned below: Name : TRUONG CONG HUY Student ID : 2270240 Place and date of born : HCMC, November 06, 1996 Address : Ho Chi Minh City With this declaring that the master thesis entitled “Analyze the impact of factors affecting rework costs in construction projects using ANN” is done by the author under supervision of the supervisors. All works, idea, and material that was gain from other reference has been cited in the correct way.
Ho Chi Minh City, , 2024 TRUONG CONG HUY vi TABLE OF CONTENTS THE TASK SHEET OF MASTER’S THESIS. iv TÓM TẮT LUẬN VĂN. v THE COMMITMENT OF AUTHOR. vi TABLE OF CONTENTS.
vii LIST OF ABBREVIATION. 2 LIST OF FIGURES. 3 LIST OF TABLES .4 Scope of Study.5 Contribution to Academic and Practical Fields .6 Outline of thesis:. 10 CHAPTER 2 LITERATURE REVIEW.1 Concepts and Definitions .2 Role construction project relevants: .3 Total investment amount for construction project .4 Cost overrun construction project definition: .5 Cost overrun construction project type: .8 Artificial Neural Network:.
27 CHAPTER 3 RESEARCH METHODOLOGIES.2 Questionnaire Survey Purpose: .3 Questionnaire Response Guideline: .1 Sample Size Identification:.2 Distribute and collect sample: .6 Data Analysis Tools: .1 Cronbach’s Alpha Coefficient (Reliability Analysis): .2 Factors ranking due to mean: .3 One sample testing: .4 One-way ANOVA: .5 Exploratory factor analysis (EFA): .1 Bartlett’s test of sphericity: .6 Artificial Neural Network (ANN):. 40 CHAPTER 4 DATA ANALYSIS .2 Data Collection Results .1 Questionnaire response result .2 Work experience result .3 Working company role result .4 Working position result .5 Number of participated project result: .6 Project type result: .7 Project scale ever participated: .8 Affection of rework on construction project: .9 Rework cost percentage on construction project:.3 Reliability analysis of scale .1 Occurrence and influence level scale reliability for Investor factor group: 51 4.2 Occurrence and Influence level scale reliability for Design factor group: .3 Occurrence level scale reliability for Supervisory factor group: .4 Occurrence level scale reliability for Contractor factor group: .5 Occurrence and influence level scale reliability for Environment and External factor group: .6 Occurrence and influence level scale reliability for Labour and Equipment factor group: .7 Occurrence and influence level scale reliability for Other group: .4 Factors ranking due to mean: .1 Mean and rank the factors according to the occurrence level: .2 Mean and rank the factors according to the influence level: .5 One sample T test implementation: .1 One-Sample T Test for occurence level: .2 One-Sample T Test for influence level: .6 Testing for the difference in mean between groups (One-way ANOVA): .1 One-way ANOVA testing for occurance level: .2 One-way ANOVA testing for influence level: .7 Exploratory factor analysis (EFA) for evaluating preliminarily measurement scale: 71 4.1 Exploratory factor analysis (EFA) for evaluating: .2 Analyze the meaning of the main components that affect the rework costs of the project:. 81 CHAPTER 5 ANN MODELLING .2 ANN Modelling parameter.2 Hidden Layer and neuron number: .3 ANN Modelling result:.4 ANN Modelling comparison:. 99 CHAPTER 6 CONCLUSION AND RECOMMENDATION .1 On the academic side: .2 On the practical side: .