2. THEORETICAL BACKGROUND
2.9 Synthesis
As presented in the literature review, the sub-process of project cost management, cost estimating, is utilized for multiple purposes during different phases of construction pro-jects and is characterized by a variety of objectives, issues, and methods.
None of the literature sources covered a full list of cost estimation methods used in the construction industry. The reason for this may be that construction as an industry is tra-ditional, and the practice has existed for a long time, which has resulted in several es-tablished methods and their variations. However, the development of information sys-tems and software applications has introduced new possible methods that likely will in-crease the efficiency of the practice but are yet to be adopted by the industry. The sum-mary of recurring cost estimation methods in the literature is presented in table 2.
Table 2. Recurring cost estimation methods in the literature.
Estimation method Description Context of study
Quantity take-off
The amount of material, labor, and equip-ment needed for completing an activity is calculated based on project drawings. Unit costs are assigned to get a cost estimate.
Apartment building projects (Günaydin & Dogan 2004), Variety of con-struction projects (Akintoye & Fizgerald 2000; Flanagan & Jewell 2018;
Chau & Long 2015; Law 1994), Estimating electrical construction ma-terials (Bryan 1991)
BIM-based QTO
Automated quantity take-off method that combines building information model and item cost information.
Apartment building projects (Firat et al. 2010), Summary of multiple BIM-researches (Wu et al. 2014), Product design of railway diesel en-gine pistons (Duverlie & Castelain 1999)
Single-rate approximation
The functional attribute of a building is mul-tiplied to provide a gross cost estimate. A lin-ear relationship between the final cost and design variable is assumed.
Variety of construction projects (Flanagan & Jewell 2018), Product de-sign of railway diesel engine pistons (Duverlie & Castelain 1999), Apartment building projects (Günaydin & Dogan 2004)
Elemental cost plan
The elemental cost plan considers the major elements of a building and provides an order of cost estimates based on the elemental breakdown.
Apartment building projects (Günaydin & Dogan 2004); Estimating electrical construction materials (Bryan 1991), Product design of rail-way diesel engine pistons (Duverlie & Castelain 1999)
Regression analysis
Statistical method for estimating the rela-tionships between the dependent variable and one or more independent variables.
Residential building projects (Kim et al. 2004; Kouskoulas & Koehn 1974; Lowe et al. 2006), Multi-storey office buildings (Karshenas 1984)
Neural network modeling
A computing system that simulates the learning process of the human brain by in-cluding a set of algorithms to recognize pat-terns.
Apartment building projects (Günaydin & Dogan 2004), Residential building projects (Kim et al. 2004; Kim et al. 2005), Highway projects (Hegazy & Ayed 1998), Concrete pavements (Adeli & Wu 1998), Multi-storey office buildings (Karshenas 1984), Variety of building projects (Emsley 2002)
Guessing, intuition
Subjective and informal method of generat-ing cost estimate based on feelgenerat-ings, personal memory, or guess.
Variety of construction projects (Akintoye & Fizgerald 2000), Estimat-ing electrical construction materials (Bryan 1991), Highway bridge construction (Chou et al. 2006)
Expert judgment
Consulting a technical expert, estimation ex-pert, or group of experts to use their under-standing and experience in generating cost estimates.
Residential building projects (Kim et al. 2004), Cost estimation models (Rush & Roy 2001)
Case-based reasoning
An analogical method for evaluating the cost of a project using past experiences from a similar set of systems.
Residential building projects (Kim et al. 2004), Pavement maintenance project (Chou 2009). Product design of railway diesel engine pistons (Duverlie & Castelain 1999), Cost estimation models (Rush & Roy 2001)
Contingency allowances
Allocated cost during planning for treating causes for contingencies, such as unplanned events.
Variety of construction projects (Flanagan & Jewell 2018), Construc-tion project bidding process (Laryea & Hughes 2011), ConstrucConstruc-tion Cost estimation principles (Carr 1989)
Range estimating
A technique that incorporates uncertainties by providing a range of estimates with their respective probabilities instead of single-point estimates.
Variety of construction projects (Akintoye & Fitzgerald 2000), Estimat-ing project cost and time in construction applications (Shaheen et al.
2007), Highway megaprojects (Molenaar 2005)
Cost estimation methods vary greatly by their characteristics, which suggests the pros and cons for companies utilizing them. The pros and cons of different cost estimation methods are presented in table 3.
Table 3. Pros and cons of cost estimation methods according to the literature.
Estimation method Pros Cons
Quantity take-off
- Provides visibility in work breakdown
- May provide extremely detailed estimates, depending on the design information available
- Easy to generate and understand
- Detailed quantity take-off is vulnerable to project scope changes (Son-mez 2008)
- Very time consuming (Bryan 1991; Duverlie & Castelain 1999)
BIM-based QTO
- Improved estimation efficiency by automated measurement (Lee et al. 2014; Wu et al. 2014)
- Less prone to human calculation errors (Firat et al. 2010) - BIM tool may be used as a data repository (Firat et al. 2010)
- Requires use and updating of associated cost databases (Lee et al. 2014) - Requires a high level of definition from design (Wu et al. 2014) - Generally, BIM tools lack standardization and provide information un-suitable for pricing (Wu et al. 2014)
- Estimates may need to be complemented by other, often subjective means to provide a finalized estimate (Firat et al. 2014)
Single-rate approxi-mation
- Straightforwardness (Flanagan & Jewell 2018)
- Possibility in generating estimate using very limited infor-mation in the early stages of the project (Chou 2009) - Possibility in providing estimating results quickly (Duverlie &
Castelain 1999)
- Assumption about the linear relationship between cost and design varia-ble is questionavaria-ble (Günaydin & Dogan 2004)
- Neglects all the other attributes of a project other than comparable pa-rameter, which may result in several adjustments (Flanagan & Jewell 2018)
- Important elements of the activity are typically neglected, and there-fore, it is difficult to justify the estimate (Duverlie & Castelain 1999)
Elemental cost plan
- Compared to quantity take-off, can be created with less final-ized design
- Successful in repetitive tasks (Bryan 1991)
- Estimates can be modified easily in case of design changes (Bryan 1991)
- The costs are presented in composite pieces, which can be re-lated to drawings (Bryan 1991)
- Relatively few rules are required to apply this method (Flana-gan & Jewell 2012)
- Assumption about the linear relationship between cost and design varia-ble is questionavaria-ble (Günaydin & Dogan 2004)
- Detailed estimates require maintained cost database or cost packages - Requires series of assumptions that are dependant on each other (Bryan 1991)
- Cost formulation may be poorly visible (Bryan 1991)
- Missing parameters require subjective modifications (Duverlie & Caste-lain 1999)
Regression analysis
- Powerful statistical tool (Kim et al. 2004)
- Can be used for both analytical and statistical work (Kim et al.
2004)
- Choosing a mathematical form of the cost function that fits the data is difficult (Karshenas 1984)
- Not appropriate when describing non-linear relationships (Kim et al.
2004)
- Defining optimal variables is tedious (Kim et al. 2004)
Neural network modeling
- May achieve extremely accurate estimates (Kim et al. 2004) - May be powerful in detecting data relationship patterns (He-gazy & Ayed 1998)
- May be trained with historical and real-time data (Hegazy &
Ayed 1998)
- Ability to model complex systems given a minimal amount of data input (Emsley 2002)
- Ability to model multi- and non-linear relationships (Günaydin & Dogan 2004; Emsley 2002)
- Knowledge acquisition process and updating the model is very time con-suming (Kim et al. 2004) and poorly visible (Hegazy & Ayed 1998) - There is no rule to determine optimal parameters for the model (Kim et al. 2004)
- It's likely that users won't be able to explain the results of the estimation (Kim et al. 2004)
Guessing, intuition - Little tedious method to provide monetary value for an activ-ity (Bryan 1991)
- Estimates are hard to defend when the cost is challenged (Bryan 1991) - Prone to biases (Lederer & Prasad 2000)
Expert judgment
- The experts may factor in differences between past project experience and requirements for new project
- Estimation may be formed with only categorical information of the project (Kim et al. 2004)
- Requires little resources in terms of time and cost (Rush &
Roy 2001)
- Prone to biases (Lederer & Prasad 2000)
-The quality of an estimate is dependant on personnel experience (Kim et al. 2004; Rush & Roy 2001)
- Method is hard to quantify
- Documenting the factors used is challenging - Estimates are black box in nature (Rush & Roy 2001)
Case-based reason-ing
- The estimate is based on actual project characteristic data - Ability to provide estimate despite lack of precise design in-formation (Chou 2009)
- CBR can be applied to qualitative and quantitative data (Chou 2009)
- Retrieving and matching past cases to new project estimate may be challenging and tedious (Kim et al. 2004; Rush & Roy 2001)
- Holistic utilization requires a CBR system and established case-base with many past cases to be effective (Duverlie & Castelain 1999; Rush & Roy 2001)
Contingency allow-ances
- Demands of accuracy and real-life representation of cost es-timates require estimating the cost of contingencies; correct use increases estimate accuracy (Carr 1989)
- The amount of risk allowance is often the result of an unstructured pro-cess (Laryea & Hughes 2011)
- In particularly competitive environments, it is difficult to use contin-gency allowances due to fear of losing the job (Smith & Bohn 1999) - Use of risk allowances is associated to uncompetitiveness of contractor (Laryea & Hughes 2011)
Range estimating
- Is found to be a useful method when quantifying uncertain-ties in high-risk activiuncertain-ties (Shaheen et al. 2007)
- Describes the probabilistic nature of future activities more re-alistically than single-point estimates (Curran 1989)
- Is not a standalone method but is used in adjunct to other methods (Curran 1989)
- Not intended for creation of highly detailed estimates (Akintoye & Fitz-gerald 2000)- Fitting statistical distributions to subjective data is difficult (Shaheen et al. 2007)- Reliable results requires a large number of simula-tion runs (Shaheen et al. 2007)
To answer the research objectives, the literature review may not be regarded as a suffi-cient source of information. In the literature, only a few references included a description of the construction cost estimation process, which was mainly limited to tendering prac-tices and could not be easily generalized. This emphasizes the need for the empirical part of this study to present the process of cost estimation activities.