İnşaat Süresini Etkileyen Faktörler Ve İnşaat Süresi Tahmin Modelleri

Abstract

Bu tez araştırması, temelde inşaat süresi konusunu ele almakta, literatürde inşaat süresini etkileyen faktörleri ve geliştirilmiş olan modelleri sistematik ve bütünsel olarak bir araya getirmeyi, ayrıca ülkemizde de kullanabilecek bir süre tahmin modelini varolanlar içinden seçmeyi ve test etmeyi amaçlamaktadır. İlk bölümde, inşaat süresi etkileyen faktörlerin belirlenmesinin ve inşaat süresinin doğru tahmin edilmesinin hem müteahhit, hem müşteri, hem de ülke ekonomisi açısından önemi vurgulanmıştır. Ayrıca süre tahmini yapmayı zorlaştıran nedenlerden söz edilmiştir. İkinci bölümde, 1970 yılından bu yana yapılan inşaat süresini etkileyen faktörler ile ilgili araştırmalar sonucu tespit edilmiş olan faktörler ele alınmış ve bu faktörler bir tablo ile özetlenmiş ve değerlendirilmiştir. Üçüncü bölümde,bu inşaat süresini etkileyen faktörlerin bir kısmının oluşturduğu, inşaat süresi tahmin modelleri incelenmiş ve modellerin geçerliliği değerlendirilmiştir. Bu modeller arasından ülkemizde kullamlabilirliği açısından diğer modellere göre daha gerçekçi sonuçlar verdiği kabul edilen Walker modeli seçilmiştir. Dördüncü bölümde ise, en gerçekçi sonuçlar verdiği kabul edilen Walker modeli detaylı olarak açıklanmış, model kullanılarak elde edilen inşaat süre performans tablosu incelenmiştir. Daha sonra, seçilen Walker'in inşaat süre tahmin modeli İstanbul'da yapılan sekiz proje üzerinde test edilmiş ve Walker modeli kullanılarak projelerin inşaat süre performans değerleri belirlenmiş ve projeler arasında performans karşılaştırması yapılmıştır. Sonuç bölümünde, Walker modelinin Türkiye' de uygulanması sırasında ortaya çıkan problemlere dikkat çekilmiş ve inşaat süresinin tahmini ile ilgili değerlendirmeler yapılmıştır.

Construction time and predicting construction time accurately, is very important for the construction industry. Therefore, in different countries, researchers have put a lot of time in determining various factors influencing construction time. They also tried to find several ways to facilitate prediction of construction time at pre-construction level. In the construction industry, construction time is usually predicted instinctively. This kind of an approach is related to the experience and the ability of the planning engineers. Therefore, in the construction industry, there are numerous methods developed in order to diminish individual's roles in decision making process of construction time prediction. Besides these methods, factors influencing construction time are also revealed through the development process. Recognising these factors is very important for both; contractor and the client. The client will have several advantages by the correct prediction of construction time such as:. Making optimum funds available to the project.. Creating a financial plan.. Procuring a solid and advantageous cash flow in pre-set time.. Choosing the correct general contractor in pre-set time. On the other hand, the advantages of the general contractor will be:. A general contractor, which predicts the construction time accurately and completes the project within the deadlines, will be rewarded in the competitiveness of the construction market.. While making construction plans, a general contractor that takes into account several factors influencing construction time will take accurate decisions during the scheduling process and therefore avoid delays during the project. Construction time and predicting construction time in advance is a key factor determining the success of a construction project. In today's competitive construction market, diminishing the construction time, without compromising of the quality and the cost, is a must in order to stay alive for a construction firm. Construction time factor, which is a key to the success in the market, is especially prominent in Turkey where the economic environment can change aspect daily. This thesis is mainly focused on construction time factor and investigates all the elements influencing construction time, unites various models of predicting construction time and evaluates them. On the other hand it also has a special motive xi Of choosing and testing the best model among them that would fit most accurately Turkey's economic conditions. On the second part of this study, these factors are shown using different tables. Researches, which are stated in these tables, are in a Chronological orders so those factors affecting construction time can be traced according to date and to the researcher's name. Various factors influencing construction time are: 1. Factors about Client / Client's representative 2. Factors about project team 3. Factors about general contractor 4. Factors about subcontractors 5. Factors about coordination of construction teams. 6. Factors about the project. 7. Factors about the construction equipment and material. 8. Factors about the environment. 9. Factors about the contract. 10. Other factors. When the table is examined, it can be concluded that researchers have not taken into account, during the early stages of their studies, the following points:. Past experiences of the general contractor.. Motivation of the construction site management.. Performance of the general contractor in its past projects.. Utilisation of modern construction technologies and resources by construction site management.. The organisational structure of construction management team and the management's decision making ability. At the same years, factors such as, project team's ability to use information technologies, Pressure on project team coming from outside of the project, workers' productivity, coordination between different teams, factors about equipment and material, politic; socio-cultural; economic limitations and legal factors were also neglected in various studies. Only after 1988 - 89 they could take an important role among the factors influencing construction time. Since 1971 during different periods various factors were discussed in these tables:. Factors concerning the project; project's complexity, absence of necessary project data, project cost, project's size, and number of floors. Factors concerning the environment; construction site's appropriateness, problems concerning the soil, construction site's accessibility, geographical formation of the construction site, weather conditions.. Factors concerning the client; Client's financial resources.. Factors concerning the general contractor; Management's control system, management's motivation role, and preparation of work schedules. On the third part of this thesis, difficulty of construction time prediction is discussed, various models developed to predict construction time are revealed and a comparison between these models is prepared in order to select most acceptable model for construction time prediction. xii Different researchers who developed a construction time prediction model are:. Bromilow Bromilow developed a simple regression model in 1980, using construction cost factor.. Ireland Ireland, alike Bromilow, developed his own simple regression model in 1983 as well as other models for construction time speed.. Mohan M.K. /Daniel WM. Mohan and Daniel used cost factor, number of floors and floor area to determine a simple regression model in 1994. On the other hand they also developed a multi-regression model through cost and floor area factors.. Walker Walker, like many other researchers, used in 1987 construction cost factor for determining construction time. How ever in 1994, he stated that only cost factor would not be sufficient for time prediction so that he offered a new model. In his new model he combined several different factors such as; project type, client's or client representative's views about the construction's quality, client representative's responsibility on construction works, coordination between the project team and the general contractor and ability of construction management team to use information technologies.. Sedef Saraç In 1995, during her thesis research at the architectural faculty of Istanbul Technical University, using various factors such as building function, construction's supporting system, construction site, site accessibility, outside wall type, number of floors, eaves height, construction floor area, gross total area, volume of earth-moving work, number of blocks etc., she developed a multi-regression model. All methods and factors used by these researchers to predict construction time, are shown on a table at the end of section three. Once this table is reviewed, cost factor can be noticed in every model. But as mentioned in section two, construction time is under influence of many factors; therefore a model based only on cost factor, could not be totally reliable. Once a comparison is made between all models of the literature, the most realistic and containing more than one variable, model turns out to be Walker's. How ever, the factors in Walker's model can not be determined before the actual start of the project, therefore the time production should be made after the initial start. In the fourth part of this research, Walker model has been explained in detail and the model introduced has been tested in 8 construction projects in Turkey. Walker has made some important researches in Melbourne University, Australia, on the factors affecting construction time and prediction of construction time. His thesis for Ph D thesis is his best research held on this subject. In this research, models developed in the literature, on the prediction of construction time, are analysed and tested on 33 projects applied on this research study. X1U 1st STEP: A data gathering system was prepared upon reviewing the literature. 2nd STEP: A questionnaire is developed in order to test data collected to prove research hypothesis. 3rd STEP: The pilot study was tested on construction managers who gave their comments on the questionnaire. 4th STEP: Replies for questionnaire and other literature studies are brought together to prepare another questionnaire for the 33 case study research. 5th STEP: Data are collected with interviews. 6th STEP: Data are analysed with statistical methods and discussed with professionals to prove the worthiness of the findings. After these steps Walker's model gave a result in form of an equation such as: Construction Time Construction Cost 0481294 ( END -VAL $0.00's) x eB B = (1.187976 x eot-act) - (0.488867 if fit-out project) + (0.105097 x obj-qual) - (0. 125269 x cr-people) + (0.079837 x cm-des-com) + (0. 104343 x cm-IT-use) e constant = 2.718281828 WORKDAYS Walker determined the construction time, which is the dependent variable as workdays. They were based on Australian construction services workdays calendar and it is assumed that there will be no work at construction sites during public holidays and weekends. l.COST VARIABLE ( END - VAL $ 000's ) Independent variable of first degree importance in Walker's model has been determined as cost variable. Construction cost is defined as the final construction cost after the deduction of wages, land cost, and headquarters' overheads are not included in construction cost. 2. EOT /ACT This variable represents the ratio of extension of construction time to the real construction duration. 4. OBJ-QUAL This variable evaluates the client representative's thoughts of construction quality. A value between 1 and 7 is requested in the questionnaire. Among these ordinal values 1 represents the minimum where 7 represents the maximum value. XIV 5. CR- PEOPLE This variable evaluates the effect of client representative's over the construction management. Ordinal values between 1 and 7 have been assigned to the questions asked and the average of these values has been calculated. Among these ordinal values 1 represents the minimum where 7 represents the maximum value. 6.CM- DES COM This variable represents the coordination between the construction management and project team. This coordination has been studied in the questionnaire in 3 stages: pre-contractual, first half of the construction and the second half. The average of the ordinal values between 1 and 7 to be assigned for each stage is calculated and this average is used in the model. 7. CM-IT- USE This variable represents the ability of the construction management team to use the information technology. Walker model has been explained in detail, the projects chosen for this research were correlated to the model and some estimated construction times have been calculated. Consequently, while the estimated construction time of some projects are found to be very close to actual ones, these are very high differences between some others. This result causes doubts on the application of Walker model in Turkey. This thesis shows Walker's construction time estimation model is much more complete compared to others. However as it is the case in other models, Walker model has some weaknesses. One of them, is that the model can not be used to estimate the construction time at the beginning stage of the construction, since the variables in the model cannot be defined before the constructions starts. How ever we cannot underestimate the importance of these variables on construction time. Therefore factors like construction management, project's organisational structure and construction quality, should be evaluated during the construction time estimation. The construction time estimation developed in the first half of the construction, is very important for all parties; because construction time may be affected by various factors at the initial start, at the first period and the second period and may be delayed at any time for any given reason. In Walker's model the factor that affects primarily the construction time is construction cost and this cost is indexed to the value of Australian $ in 1990. But in Walker's research, the model is not appropriate for converting Australian $ to other currencies or implementing it on countries with a high inflation rate such as Turkey.