Please use this identifier to cite or link to this item: http://hdl.handle.net/11527/13948
Title: Genetik Algoritma Ve Monte Carlo Simülasyonu İle Bir İnşaat Projesinde Alt Yüklenici Seçimine İlişkin Süre Maliyet Kalite Optimizasyonu Ve Risk Değerlendirmesi
Other Titles: Time Cost Quality Optimization And Risk Evaluation About Selecting Subcontractors In A Construction Project By Using Genetic Algorithm And Monte Carlo Simulation
Authors: Tatar, Gül Polat
Kaplan, Barış
10039175
İnşaat Mühendisliği
Civil Engineering
Keywords: Genetik Algoritma
Monte Carlo Simülasyonu
İnşaat Projelerinde Alt Yüklenici Seçimi
Süre Maliyet Kalite Optimizasyonu
Risk Değerlendirmesi
Genetic Algorithm
Monte Carlo Simulation
Selecting Subcontractors In Construction Projects
Time Cost Quality Optimization
Risk Evaluation
Issue Date: 9-Jun-2014
Publisher: Fen Bilimleri Enstitüsü
Institute of Science and Technology
Abstract: İnşaat sektöründe alt yüklenici kullanımı, yaygın olarak uygulanmaktadır. Çeşitli sorunları da beraberinde getiren bu sistemde, alt yüklenici seçimine dikkat edilmeli ve alt yüklenici yönetimine özen gösterilmelidir. İnşaat firmaları, günümüzün rekabet koşulları altında inşaat projelerini; en kısa sürede, en az maliyetle ve mümkün olan en kaliteli şekilde tamamlamaya çalışmaktadır. Bu da, bu kıstasların optimize edilmesini gerektirmektedir. Ayrıca, inşaat sektörünün barındırdığı belirsizlik ve riskler iyi analiz edilmeli ve risklere karşı izlenecek stratejiler belirlenmelidir. Bu çalışmada; Genetik Algoritma ve Monte Carlo Simülasyonu yöntemleri kullanılarak gerçek bir lojistik merkezi inşaatı projesinde, alt yüklenici seçimine ilişkin bir süre maliyet kalite optimizasyonu ve risk değerlendirmesi modeli oluşturulmuştur. İlk bölümde; tez çalışması hakkında özet bilgi verilmiş ve çalışmanın amacı, kapsamı, yöntemi ve içeriği ile ilgili genel bilgi verilmiştir. İkinci bölümde; inşaat sektöründe alt yüklenici seçimi olgusu ele alınmıştır. İnşaat sektöründe alt yüklenici kullanımı ile ilgili bilgi verilmiştir. Alt yüklenici kullanılmasının sebepleri aktarılmıştır. Alt yüklenicilerin sorumlulukları anlatılmış ve alt yüklenicilik sisteminin işleyişi açıklanmıştır. Alt yüklenici kullanımının getirdiği sorunlar üzerinde durulmuş ve alt yüklenici yönetimi ile ilgili uyarılarda bulunulmuştur. Alt yüklenici seçimi ile ilgili yapılmış çalışmalar özetlenmiştir. Üçüncü bölümde; inşaat sektöründe süre, maliyet ve kalite yönetiminin önemi vurgulanmıştır. İnşaat projelerinde gözlemlenen ve tezin vaka çalışması bölümünde kullanılan maliyet tipleri açıklanmıştır. Dördüncü bölümde; inşaat sektöründe karşılaşılan belirsizlikler ve riskler anlatılmıştır. Risk yönetimi kapsamında; risklerin belirlenmesi ve analiz edilmesi ile, risklere karşı geliştirilecek olan stratejiler hakkında bilgi verilmiştir. Beşinci bölümde; Genetik Algoritmanın tarihçesi, uygulama alanları ve faydalı olduğu koşullar özetlenmiştir. Genetik Algoritmanın uygulama metodolojisi ve aşamaları gösterilerek, yöntemin avantaj ve dezavantajları belirtilmiştir. İnşaat mühendisliği alanında GA ile ilgili yapılmış çalışmalar sıralanmıştır. Altıncı bölümde, Monte Carlo Simülasyonu yönteminin tarihsel gelişimi aktarılmış, yöntemin uygulama sahaları hakkında bilgi verilmiştir. Yöntemin kullanılmasını gerektiren sebepler ve uygulama detayları açıklanmıştır. MCS’nun güçlü ve zayıf yönleri belirtilerek, inşaat sektöründeki kullanımları özetlenmiştir. Yedinci bölümde; gerçek bir inşaat projesine ilişkin vaka çalışmasına yer verilmiştir. Çalışmanın kapsamı hakkında genel bilgi verilerek, modelin nasıl oluşturulduğu açıklanmıştır. Aktivite, iş programı ve proje verileri kabulleri kurgusu yapılmış; temin edilen alt yüklenici verileri, belirtilen yazılımlara girilerek sonuçlar değerlendirilmiştir. Son bölümde; yapılan çalışmaya ilişkin sonuç ve önerilere yer verilmiştir. Projede kullanılması gereken optimum alt yüklenici diziliminin belirlenmesinde, GA ile çalışan Evolver yazılımı kullanılmıştır. Optimizasyon çalışmasında yer alan süre, maliyet ve kalite verilerine, MCS ile çalışan @Risk yazılımı yardımıyla risk faktörü etkisi eklenmiştir. Simülasyon sonuçlarında; toplam proje süresi, maliyeti ve kalitesi değerlerinde olumlu ve olumsuz sapmalar olabileceği gözlemlenmiştir. Elde edilen sonuçların, alt yüklenici seçimi ile ilgili farklı bir yaklaşım oluşturması beklenmektedir. Süre, maliyet ve kalite kıstasları arasındaki değişimlerin daha iyi yorumlanacağı düşünülmektedir. Risk ve belirsizliklere karşı önlem alınması için, bir farkındalık oluşturulması amaçlanmıştır. GA ve MCS yöntemleri ile çalışan sektörel yazılımların; gerçek bir inşaat projesi optimizasyonu ve risk değerlendirmesi modelinde kullanılmasının, Türk inşaat sektörü akademik literatürüne katkıda bulunacağı düşünülmüştür. Bu çalışma; aktivite listesinin genişletilmesi, daha fazla alt yüklenici alternatifi eklenmesi, karar verme kıstaslarının arttırılması ve çözümlerin daha gelişmiş yazılımlarla yapılması sayesinde ileriki akademik araştırmalarda daha da geliştirilebilecektir.
Contractors choose to work with subcontractors in most of the construction projects because of various reasons. A good management of the subcontractors which have several responsibilities in different work items, is an important issue for the success of the project. Although using subcontractors bring some problems, contractors have the chance to reduce these problems when they select the right companies. Completing the construction projects; on time, within the specified budget and at a required quality level is the owners’ main demand. It is getting more important to optimize time, cost and quality criteria under the competitive market conditions of the recent years. Contractors should be careful about these criteria when selecting subcontractors which work as solution partners during the most of the project period. In addition, risk factors should be considered and a risk management strategy should be built in construction projects which have so many stakeholders, uncertainties and activities. In this research; by using Genetic Algorithm and Monte Carlo Simulation methods, a time cost quality optimization and risk evaluation model about selecting subcontractors in a real construction project is configured. In determining the research subject process, the suitability of the subject for the professional business issues is considered. During the literature searching, many contemporary studies about the subject have been found. A case study has been taken about a real construction project. So, research constraints are minimized and the capacity for gathering data is maximized. In determining the scope, level of detail is optimized for the aim of the study. For example; to reduce the number of the activities, subcontractor based work packages are used instead of detailed work program which is made of hundreds of activities. Data for time, cost and quality are assigned to these work packages. Modelling which is one of the ampiric research methods is chosen as a methodology. A real construction project in Samandıra İstanbul is used for modelling. Examining the company documents and making an interview with the technical office manager of the company are implemented as data gathering methods. In the first chapter, a brief information about the research is given. The aim, scope and method of the research are described. In the second chapter, subcontractor selection in construction industry is considered. Information about the usage of subcontractors in construction industry is given. The reasons for using subcontractors are sorted as; to reduce the overhead, construction, material and investment costs, to decrease the uncertainty and risk level, to minimize the supervision problems and the flexibility to work in different countries. The responsibilities of subcontractors are sorted as; supplying materials and machines, implementing new technology, work standardization, cooperation with other shareholders and sharing the technical knowledge. The phases of subcontracting system are explained as; tendering phase, contracting phase, construction-supervision phase and progress payment-final account phase. The problems brought by using subcontractors are examined as; loss of supervision, new uncertainities, time problems, sub-subcontractors, quality failures, uncompleted activities, unsuitable material usage and material damages and losses. The principles of subcontractor management are described as; making prequalification evaluation, clarifying the demands, privacy of tenders, well-balanced contracts, making the work program together, knowing their staff, fair relationships, weekly meetings, periodic performance evaluation, attention to hygiene, maintenance and work safety and paying them on the right time. Previous researches about subcontractor selection in construction industry are summarized. In the third chapter, the importance of time management, cost management and quality management in construction industry is emphasized. Types of costs in construction projects which are also used in the case study of the research are given as; direct and indirect costs. Time cost quality optimization is explained and new contracting methods via this issue are shown. Previous researches about time cost quality optimization in construction industry are summarized. In the fourth chapter, the terms of uncertainty and risk are given. The reasons of the uncertainties and the risks encountered in construction industry are described as; having big budgets and many cost codes, long time projects, many shareholders and complex organizations, large work programs made up of many activities, natural conditions, critical quality levels and vulnerability to cultural, political and legal effects. Tendering risks, political risks, market risks, financial risks, natural risks, cultural risks, company risks, construction risks, design risks, contract risks, legal risks and environmental risks are given as the types of risks in construction industry. Risk management process is explained with the following phases; identification of risks, evaluation of risks, analysis of risks and generating a risk attitude. Risk analysis methods are listed. The strategies which should be developed against risks are sorted as; avoiding risks, reducing risks, accepting risks and transferring risks. In the fifth chapter; the history of Genetic Algorithm method is explained. Fields of application are listed as; optimization problems, programming and information systems, mechanical learning applications, econmical and social systems modelling, financial models, marketing strategies and production problems. The conditions in which Genetic Algorithm method is used are summarized as; when the research field is large and the subject is complex, when the data about the subject is insufficient, when the solution is difficult by deterministic methods, when the common software products are insufficient and when the data for modelling can not be reached. The terms which are used in Genetic Algorithm are described. The methodology and implementation steps of Genetic Algorithm are presented as; generating the initial population, choosing the parent chromosomes, reproduction, cross-over, mutation and termination. The advantages and disadvantages of the method are described. Previous researches about Genetic Algorithm in the field of civil engineering are given. In the sixth chapter, general information about Monte Carlo Simulation method is presented. Historical development of Monte Carlo Simulation method is described. Fields of application of the method are explained as; manufacturing systems design, assembly line balancing, work force planning, material transportation systems, generating new military weapons and tactics, purchase order planning in inventory systems, design of communication systems and message protocols, design and operation of highways, airports, metro lines and harbours, determining the number and the locations of the ambulances, determining the locations of the fire stations and the number of their equipments, analysis of financial and economical systems, design of distribution channels, detection of the hardware and softare necessitations for the computer systems, management games and space fly simulations. The reasons for using Monte Carlo Simulation and the details of the implementation of the method are summarized. Strengths and weaknesses of it presented and previous studies in the field of civil engineering are given. In the seventh chapter, there is a case study about a real logistic center construction project owned by one of the biggest healthcare holdings of Turkey. A time cost quality optimization and risk evaluation model about selecting subcontractors by using Genetic Algorithm and Monte Carlo Simulation is generated. General information about the project is presented. To gather time, cost and quality data of the subcontractors of the project, the documents of the company are examined and an interview is set with the technical office group manager. Main work packages of the master work schedule of the project are determined. Relationships between the activities are listed and CPM arrow diagram are created. Some assumptions are made to complete the model. To find the optimal subcontractor combinations under different constraints by Genetic Algorithm, Evolver software of The Decision Tools Suite by Palisade Corporation is used. To add the risk effect to the problem, @Risk software of the same company is used. In the last chapter, conclusions and recommendations are given. It is seen that Evolver software can be used to find the optimal subcontractor combination for the activities of the project. The software has generated an optimal solution among thousands of combinations which can not be solved manually. The important point is to describe the project constraints correctly. There is not a solution in which the results for time, cost and quality are all advantageous, so the project manager should make a decision among the possible solutions. At the beginning of construction projects; total project duration is given in the work schedule, total project cost is calculated in the project budget and quality acceptance criteria is described in technical specifications. However, the results for these criteria can be different at the end of the project. For this reason, a risk effect is added to the model by using @Risk software. According to simulation results, it is seen that total time, cost and quality values can deviate from their initial values. From the point of construction industry, it is expected that the results of this research will help contractors to set up an effective subcontractor selection system. Furthermore, the model will give a chance to contractors to understand the tradeoff between time, cost and quality. Another aim of the study is to increase the awareness level of the companies about uncertainties and risks. In terms of academic field, Genetic Algorithm and Monte Carlo Simulation methods are implemented in a time cost quality optimization and risk evaluation problem. A multi criteria decision making model is created by using industrial software products. It is thought that this research will be contributional to Turkish construction industry literature. The model generated in this research can be improved in further academic studies by increasing the number of activities, the number of subcontractor alternatives and the number of decision criteria. Using more sophisticated software products will be advisable as the technology level increases.
Description: Tez (Yüksek Lisans) -- İstanbul Teknik Üniversitesi, Fen Bilimleri Enstitüsü, 2014
Thesis (M.Sc.) -- İstanbul Technical University, Instıtute of Science and Technology, 2014
URI: http://hdl.handle.net/11527/13948
Appears in Collections:İnşaat Mühendisliği Lisansüstü Programı - Yüksek Lisans

Files in This Item:
File Description SizeFormat 
10039175.pdf1.6 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.