MULTIPLE CRITERIA SELECTION OF PILE-COLUMN CONSTRUCTION TECHNOLOGY

Edmundas Kazimieras Zavadskas, Saulius Sušinskas, Alfonsas Daniūnas, Zenonas Turskis, Henrikas Sivilevičius 1, 3, 4, 5Vilnius Gediminas Technical University, Saulėtekio al. 11, 10223 Vilnius, Lithuania 2Kaunas University of Technology Panevėžys Institute, S. Daukanto g. 12-138, 37164 Panevėžys, Lithuania E-mails: 1edmundas.zavadskas@vgtu.lt; 2saulius.susinskas@ktu.lt; 3alfonsas.daniunas@vgtu.lt; 4zenonas.turskis@vgtu.lt (corresponding author); 5henrikas.sivilevicius@vgtu.lt Received 20 Dec. 2010; accepted 04 Nov. 2011


Introduction
In construction, piles can be used in various ways. In urban areas, many high-rise buildings and viaducts are founded on a pile foundation. Construction technologies are highly dependent on in-situ conditions, e.g. soil conditions are particularly important for a foundation. The way the designed and actual founding depths of foundations correspond to variability of geological conditions has long been a concern (Zhang et al. 2011b). Tomlinson and Woodward (2008) presented a lot of pile design examples. Sivilevičius et al. (2012) presented results of an experimental study on technological indicators of pilecolumns at a construction site. Based on in-situ investigation of natural soil conditions, regression equations have been determined, which can be very useful when planning similar works at a construction site. Besides, they allow determining duration and energy consumption of construction works. Zhang and Dasaka (2010) evaluated the spatial variability characteristics at a weathered soil site. Sušinskas et al. (2011) presented the process for selection of the most fitting and effective pile-column instalment alternative. The model is based on ARAS method and AHP technique. Zhang et al. (2011a) proposed a two-stage analysis method to study the behaviour of pile groups with rigid elevated caps. A single pile foundation utilizes a single, generally a large-diameter structural element to support all of the loads (weight, wind, etc.) of a large above-surface structure. Yoon et al. (2011) presented the evaluation results of the load test on columns and the rationale used for the selection of the resistance factor. Zhao et al. (2009) presented the model for stability analysis of high pile-column bridge pier. Zhang et al. (2011b) analysed excavation-induced responses of loaded pile foundations considering the uploading effect. Zhao et al. (2007) revisited the stability analysis regarding the pile-columns of a bridge pier.
Sustainable development aims to reconcile economic growth, social progress and frugal use of natural resources, to maintain ecological balance and to ensure favourable living conditions for current and future generations (Raslanas et al. 2011). Selection of an investment strategy and related decision making relies heavily on personal experience and behaviour (Wu et al. 2012;Šaparauskas et al. 2011;Banaitienė et al. 2011). Multiple criteria decision making is an important part of modern decision science Zavadskas et al. 2008). How to select an effective algorithm for a multiclass classification task is an important yet difficult issue (Peng et al. 2011). Most of the real-world multiple criteria decisionmaking problems contain a mixture of quantitative and qualitative criteria (Nieto-Morote, Ruz-Vila 2011; Kaklauskas et al. 2011;Merigo, Gil-Laufente 2011). The typical MCDM problem is concerned with the task of ranking. In order to evaluate the overall efficiency of technological alternatives, typically it is necessary: a) to identify the system for evaluation of criteria that relates the system capabilities to goals; b) to develop alternative systems for attaining the goals (generating alternatives); c) to assess a finite number of decision alternatives, each of which is described in terms of different decision criteria which are taken into account simultaneously; d) to apply a normative multiple criteria analysis method; e) to accept one alternative as the most preferable; f) to gather new information and go into the next iteration of multiple criteria optimization if the final solution is not accepted.
At the beginning of his book, Zeleny (1982) stated that "It has become more and more difficult to see the world around us in a unidimensional way and to use only a single criterion when judging what we see". In reality, the modelling of engineering problems is based on a different kind of logic taking into consideration the existence of multiple criteria, the conflicting aims of decision maker, the complex, subjective and different nature of the evaluation process, and the participation of several decision makers. The use of the new and modernisation of the existing technologies as well as the selection of the most suitable alternative among those feasible with the help of different models are challenging tasks for the modern civil engineering Krayushkina et al. 2012). Estimation and modelling of problems depends the recent advances achieved in different fields (Dzemyda, Sakalauskas 2011). Selection of the right construction technology plays a vital role in the overall performance of a project, thus posing the most crucial challenge for any contractor. Numerous and often conflicting objectives and alternatives, such as tender price, completion date, and experience, need to be considered. Recently, to assist contractors and stakeholders in decision-making, there has been a trend to move away from the "lowest-price wins" principle and subjective judgement to the multiple criteria selection approach in the selection of alternatives (San Cristóbal 2012).

Case study
Projects with pile-columns are complex systems that are rather difficult to select in practice. For this reason, a decision-maker should possess a large amount of multidisciplinary knowledge and be familiar with multidisciplinary techniques of operations research. The case study presents the process of selecting the pile-column alternative for a building that stands on the aquiferous soil. The aim of the study is to design and install the most effective pile-columns. The study shows how a decision-maker can find the most reasonable alternative with the help of a certain dataset. Taking into account the aforementioned suggestions and references of experts as well as the aim to install the most effective pile-columns, the five following alternatives were considered (Table 1). Driving the reinforced concrete ring using a punch, driving the pole, construction, positioning and adjustment of the mounting jig for the column, placing in situ concrete, and column mounting.
Driving the reinforced concrete ring Driving the pole Positioning and adjusting the mounting jig Column mounting a 2 Driving the reinforced concrete ring by applying a punch, driving the pile, placing in situ concrete mixture basement with a nest for the column mounting, and column mounting.
Driving the reinforced concrete ring Driving the pole Positioning and adjusting the mounting jig Column mounting Continue of Table 1 Alternative Short description of the alternative a 3 Driving the steel ring by applying a punch, driving the pile, placing in situ concrete mixture basement with a nest for the column mounting, and column mounting.
Driving the reinforced concrete ring Driving the pole Positioning and adjusting the mounting jig Column mounting a 4 Driving the steel ring by applying a punch, driving the pile, placing in situ concrete basement with a nest for the column mounting, removing the steel ring, and column mounting.
Driving the reinforced concrete ring Driving the pole Positioning and adjusting the mounting jig Column mounting a 5 Drilling the leader bore with 0.8 m in diameter and 1.0 m in height, driving the reinforced concrete ring, driving the pile, positioning and adjusting the mounting jig for the column, placing in situ concrete mixture, and column mounting.
Driving the reinforced concrete ring Driving the pole Placing in situ concrete basement with a nest for the column Column mounting

Remark
The inner diameter of all driven rings equals to 1.0 m.
The construction technology of alternatives is described by six criteria. The set of criteria was determined by qualified civil engineers and shown in Table 2. The selection is based on a set of criteria: labour expenditures (x 1, hours), cost of instalment (x 2 , €), consumption of concrete (x 3 , m 3 ), consumption of steel (x 4 , kg), machinery expenditures (x 5 , hours), and consumption of energy (x 6 , GJ). The criteria set for evaluation is selected considering the factors that influence the efficiency of the construction process. Significance of criteria significances (weights) was determined with the help of the expert judgement method and the analytic hierarchy process (AHP) method. Integrated criteria weights were applied in the solution process.  (Kendall 1970) at the first stage of criteria weight determination. Zavadskas et al. (2010a) provided a detailed presentation of the algorithm and discussed peculiarities of weight determination. The weights p j of attributes presented in Table 1 were determined by application of the expert judgment method proposed by Kendall. This expert judgment method was implemented at the following stages: a) calculation of values t; b) calculation of weights w; c) calculation of values S; d) calculation of values T k ; e) calculation of concordance value W; f) calculation of values χ 2 ; g) testing the statement χ 2 > χ 2 tbl . The values t jk for statistical processing were obtained by interviewing the respondents. Kendall (1970) has demonstrated that, when n > 7, the value χ 2 α,ν = W·r·(n -1) has a distribution with degrees of freedom ν = n -1, where n is the number of attributes considered and r -the number of experts. If the calculated value χ 2 is larger than the critical tabular value tbl 2 χ for the pre-selected level of significance α, then the hypothesis about the agreement of independent expert judgments is not rejected. In the case study, the number of experts r = 26, the degrees of freedom ν = n -1 = 5 and the pre-selected level of significance is α = 0.05. The calculated concordance coefficient based on the weights of attributes is W = 0.558. The tabular value 2 15.08 ( 0.05) tbl χ = α = (Fisher, Yates 1963 then the assumption is made that the coefficient of concordance is significant and expert rankings are in concordance with 95% probability. During the next step, experts applied the WEAR software (which contains the AHP method) to determine criteria weights (Zavadskas et al. 2012) (see Table 3).
In decision analysis, the analytical hierarchy process (AHP) and the analytical network process (ANP) are widely used to assess the key factors and analyse the impacts and preferences of decision alternatives (Ergu et al. 2011a, b).
The recent developments of decision making models based on the AHP (Saaty 1980;Saaty, Zoffer 2011;Vaidogas, Sakenaite 2011)   Integrated criteria weights were calculated during the third stage of criteria weight determination (Table 4). Three different multiple criteria decision making methods -TOPSIS, COPRAS and ARAS -were selected to solve the investigated problem An Additive Ratio Assessment (ARAS) method ) is based on the argument that complicated phenomena could to be understood by using simple relative comparisons. It is argued that the ratio of the sum of normalised and weighted values of criteria, which describe an alternative under consideration, to the sum of the values of normalised and weighted criteria, which describes the optimal alternative, is the degree of optimality, which is reached by the alternative under comparison. The recent developments of decision making models based on the ARAS method are listed below: Keršulienė and Turskis (2011) presented an integrated fuzzy multiple criteria decision making model for the selection of an architect; Turskis and Zavadskas (2010b) performed multiple criteria analysis in order to select the location for a logistics centres; and Zavadskas et al. (2010b) analysed foundation alternatives.
The method of complex proportional assessment COPRAS (Zavadskas, Kaklauskas 1996) assumes direct and proportional dependence of significance and utility degree of investigated alternatives on a system of criteria adequately describing the alternatives, and on values and weights of the criteria. This method was used to solve various problems in construction.
The recent developments of decision making models based on COPRAS methods (Podvezko 2011)  The TOPSIS (Technique for Order Preference by Similarity to Ideal Solution) method determines a solution with the shortest distance from the ideal solution and the farthest distance from the negative-ideal solution (Hwang, Yoon 1981). Kalibatas et al. (2011) used it in order to solve the problem of the assessment of dwellinghouses, determining the ideal indoor environment. Rudzianskaite-Kvaraciejiene et al. (2010) evaluated the effectiveness of road investment projects.
The description of the methods is presented in Table 5.
First of all, the initial decision making matrix was prepared. The problem was solved by applying three different multiple criteria decision making methods: TOPSIS, COPRAS and ARAS. The solution process of the problem is presented in Table 6. Table 5. Description of TOPSIS, COPRAS and ARAS methods TOPSIS COPRAS ARAS m -number of alternatives, n -number of criteria describing each alternative, x ij -value representing the performance value of the i alternative in terms of the j criterion.   Table 5 TOPSIS COPRAS ARAS Normalisation of the initial decision making matrix ( ) ∑ Utility degree Ranking of alternatives preferable most the is K i i , max

Conclusions
Overall, the main advantages that the MCDM provides in decision making could be summarized in the following aspects: the possibility to analyse complex problems; the possibility to aggregate both quantitative and qualitative criteria in the evaluation process; good evidence of decisions; the option for a decision-maker to participate actively in the decision-making process; and the use of flexible scientific methods in the decision making process. According to the newly proposed model, the priorities of alternatives can be determined according to the utility function value. Consequently, it is convenient to evaluate and rank decision alternatives when this model is used.
The degree of the alternative utility is determined by comparison of the analysed variant with ideally the best one.
It can be stated that the ratio with an optimal alternative may be used in cases when it is required to rank alternatives and find ways to improve alternative projects.
Three MCDM methods were applied. Alternatives according to all methods rank in the same way: This means that the most preferable alternative is a 4 that must be selected and implemented.
The proposed model can be modified and applied to solve different problems: to select, assess and rank constructions, technologies and other alternatives.