Reliability Assessment of Research into Contractors ... - ScienceDirect

Available online at www.sciencedirect.com

ScienceDirect Procedia Engineering 122 (2015) 251 – 257

Operational Research in Sustainable Development and Civil Engineering - meeting of EURO working group and 15th German-Lithuanian-Polish colloquium (ORSDCE 2015)

Reliability Assessment of Research into Contractors’ Bidding Decisions Agnieszka Leśniaka* b

Cracow University of Technology, ul. Warszawska 24, Kraków, 31-155 Poland

Abstract The most popular system for contracting construction works in the Polish market is the tender procedure. The selection of tenders in which a company will take part is very important as it plays a vital role in building its position on the market. Making a bid/no bid decision is influenced by a number of factors. The research carried out so far has been concerned mainly with identification of various kinds of factors influencing contractors’ bidding decisions. The present study describes a group of factors – identified using a survey method – that may influence the contractor’s decision to participate in a tender procedure. An important problem revealed by the survey is assessment of experts’ opinions consistency. The method used in this paper for the consistency evaluation is Kendall's coefficient of concordance. ©©2015 Published by Elsevier Ltd. This 2015The TheAuthors. Authors. Published by Elsevier Ltd. is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). committee of the Operational Research in Sustainable Development and Civil Peer-review under responsibility of the organizing Peer-review under responsibility of the organizing committee of the Operational Research in Sustainable Development and Civil Engineering- meeting - meeting of EURO working and German-Lithuanian-Polish 15th German-Lithuanian-Polish colloquium. Engineering of EURO working groupgroup and 15th colloquium Keywords: contractors’ bidding decision, bid/no bid factors, reliability assessment

1. Introduction The ability to select the appropriate contracts may not only determine the overall health and success of the company, but even its survival. Participation in tenders for projects, which are not suitable for the company can

* Corresponding author. Tel.: +48-12-628-23-09; fax: +48 12 628-30-93 E-mail address: [email protected]

1877-7058 © 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license

(http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of the Operational Research in Sustainable Development and Civil Engineering - meeting of EURO working group and 15th German-Lithuanian-Polish colloquium

doi:10.1016/j.proeng.2015.10.033

252

Agnieszka Leśniak / Procedia Engineering 122 (2015) 251 – 257

result in big losses of both money and time. On the other hand, not participating in a tender may be a failure to take advantage of an opportunity to make a profit, to improve the company's position and strength in the market and establish relationships with new clients. Taking the decision to participate in a tender procedure is influenced by numerous factors related to the company, its environment and the undertaking referred to in the tender. First studies of the factors influencing bidding decisions were done among construction works contractors in the USA in 1988 [1], another study was carried out in Great Britain [13], followed by one in Syria [16], Singapore [4], Saudi Arabia [2]. The authors of these works have observed that although the research was done in different countries, the factors featured at the top of the ranking lists are more or less the same. Nevertheless, there is a group of factors in each country which are characteristic only of the specific local market. These observations suggest that the factors influencing bidding decisions are to a considerable extent dependent on the market and the environment in which a company operates. Similar studies have been carried out in Poland with the use of the survey method [8,9]. However, they may have a deficiency, which is the quality of the collected information. It is therefore important to verify the collected answers. The article presents briefly the research results and the reliability assessment of the obtained opinions using the Kendall-Smith coefficient of concordance. 2. Awarding construction works contracts in the construction market Methods of soliciting contractors in the construction market depend on the type of market, the type of contract and its value. In the case of the public sector in Poland, the contracting authority is obligated to observe the Public Procurement Law (The Act of 29 January 2004) [15]. The Act specifies eight types of public procurement procedures: 1. open tendering, 2. restricted tendering, 3. negotiated procedure with publication, 4. negotiated procedure without publication, 5. competitive dialogue, 6. single-source procurement procedure, 7. electronic bidding, 8. request for quote. This type of procurement is the only one out of eight listed above which does not apply to the award of public contracts for construction works. The choice among them depends on the conditions and restrictions introduced in individual situations. According to article 2 (The Act of 29 January 2004), construction works should be understood to mean either execution alone or design and execution of construction works specified in the regulations issued pursuant to article 2c, or a civil structure as well as the implementation of a civil structure by any means, in accordance with the requirements specified by the contracting authority. In compliance with the above regulations, a public investor may implement a project in two ways: in the Design-Bid-Build system or in the Design & Build system. In the first one, it is obliged to award two separate contracts – one for the service of design documentation preparation and the other for construction works. In the second one – it will entrust one contractor with designing the structure and with the construction works. According to data obtained from the Public Procurement Bulletin, 46536 construction works contracts were awarded in 2014. The value is comparable to the ones in previous years (Fig. 1). The prevailing type of contracts awarded in recent years are contracts for construction works. Joint contracts for both designing and building related to it make up a small percentage. As the research presented in [10] suggests, it reaches approximately 1 – 2 % of all contracts annually. The most frequently used mode of selecting the contractor for the works is open tendering. It was applied in 86 % of all tendering procedures in 2014. Generally, it is the most frequently used mode in the sector of public procurement, works, services and supplies alike (Table 1) .

253


48957 50000

Number of contracts awarded

45000

47764

46536

41708

40000 35000 30000 25000 20000 15000 10000 5000 0

2011

2012

2013

2014

Fig.1 The number of public contracts for construction works awarded in recent years. Source: the author on the basis of the data obtained from [11]. Table 1. The percentage of public contracts awarded with the use of tendering procedure. Source: [12] Procedure Open tender

2011 82.07%

Years / percentage of contracts 2012 84.43%

2013 81.00%

Open tendering is a bidding process where all economic operators interested in responding to a public contract notice are invited to submit their tenders (Art. 39 of the PPL). In contrast, restricted tendering is a mode in which, in response to a public notice, contractors submit applications to participate in bidding, whilst only those invited to tender may submit their bids (in compliance with Art. 47 of the PPL). Using the open tender procedure has certain important advantages: - no requirement to state reasons and no need of any evidence (art. 39 of the PPL), - it is the most competitive type of procurement - bids may be submitted by all interested contractors, - the procedure is not complicated. The contracting authority announces a tender, in response all the interested parties submit bids, and the most advantageous bid is selected, - procedural time is the shortest when compared to other types of procurement (in 2013, the aver-age duration of this procedure was 38 days, while in the use of restricted bidding the duration is 81 days, and in the case of competitive dialogue as long as 180 days [12]. In the private sector, the client decides what form of search for a contractor is the most advantageous. No piece of legislation imposes on a private party the need to publish a contract notice in a particular place and a particular way (regardless of the value of the contract). The employer may, for example, provide information about the tender to the public by posting the notice at its registered office, on its website or by using web portals. It can also inform only selected contractors about the tender, by sending them an invitation to tender. In each of the listed types of tender procedures in the construction industry (construction works, supplies and services), the contracting entity may start the procedure by issuing a call for tenders. A potential contractor must take the decision whether to take part in the tender or not and then initiate the process of preparing the bid. Appropriate performed the decision-making process can lower the risk of investing and execution of works [5,6,7,17]. 3. Factors influencing bidding decisions - contractors' opinions A detailed description and results of research conducted in Poland can be found in the study based on a survey implemented by [8] in which a large group of Polish contractors were invited to participate. In order to collect relevant information from contractors a questionnaire consisting of two parts was designed. Part one included questions concerning the company, thus allowing to create a brief description of the contractors taking part in the study. Part two consisted of questions regarding the degree of the influence of the proposed factors on the bid/no bid decision. The questionnaire was sent to 150 companies belonging to micro-, small- and medium-sized construction

254


businesses. 61 of them (out of 160, that is 38%) responded to a questionnaire. Among the responses, 38% of companies signed more than 75% of contracts resulting from bidding, and only 19% of respondents signed no more than 25%. Contractors task was to specify the degree of importance of 16 factors, marking them on a 1-7 scale, where 1 was the factor with no influence on the decision and 7 was the one with the greatest significance in decision making. Then, for each of the factors, average results were calculated (1) thus giving the response a particular rank, which allowed to create a ranking of the proposed factors (Fig.2).

Fig.2. Factors influencing bidding decisions as seen by contractors. Source: [8] Ni

¦a Si

j 1

Ni

ij

for

i 1,2... 5

where: Si – average for the i-th factor Ni – total number of responses for the i-th factor; aij– assessment degree assigned to the i-th factor in the j-th response.

(1)

255


The survey revealed that the three key factors affecting the decision to participate in a tender for construction works were the following: 1. the type of works, 2. past experience with similar projects, 3. contract conditions (for instance, the type of remuneration: a lump sum or unit price contract, the possibility of obtaining an advance payment, time for completing the works, or the amount of performance security). On the other hand, the factors that had the least impact on the bidding decisions, in the opinion of contractors, were: 1. the possibility of subcontracting, 2. the need for special equipment, 3. the difficulty of the work. 3.1. The collected opinions consistency assessment The most frequently used tool for researching the factors affecting bidding decisions so far has been the survey. One advantage of the survey methodology is its relatively low cost, another – the possibility of obtaining answers from a large number of respondents in a short time. A disadvantage may be the quality of the collected information, which often lacks objectivity. Therefore, assessment of the experts’ opinions consistency is an important issue. If the level of consistency of the opinions expressed by a group of experts is adequately high, they may give grounds for formulating certain general opinions. Finding the opinions inconsistent prompts taking some action aimed at eliminating the source of inconsistency, or – if the source cannot be removed – refraining from formulation of a general opinion [3]. The Kendall-Smith coefficient of concordance, also called Kendall’s coefficient W, has been used in assessing the experts’ opinions consistency. Its values range from 0 to 1. The higher the value, the greater the consistency of the experts’ opinions. The coefficient of concordance adopts the following form [3]:

ൌ

ͳʹ

ʹ ሺ͵ Ǧሻ

(2)

where: m – the number of experts, n – the number of variants, the number of factors rated by experts. ௡

௠

ܵ ൌ ෍ሺ෍ ‫ݔ‬௜௝ െ ‫ݔ‬ҧ ሻଶ

(3)

௝ୀଵ ௜ୀଵ ௠

௡

ͳ ‫ݔ‬ҧ ൌ ෍ ෍ ‫ݔ‬௜௝ ݊

(4)

௜ୀଵ ௝ୀଵ

where: x ̅ – mean value of the sum of ranks for all variants (factors), xij – rating of the i-th expert for factor j. The above approach is applied in the so-called strong ordering, where all the numbers indicating the positions of the rated factors are different. In the situation where experts are asked to rate n factors on the scale from 1 to k, where k is a natural number, but less than n, a part of the factors will have to be tied by the same rank. We are then faced with weak ordering [3]. If the ratings done by experts yield series with tied ranks, the consistency assessment may be carried out with the use of the coefficient of concordance W determined by the following formula [3]:

256


ܹൌ

ܵ ݉‫ܩ‬

(5)

where: ௠ǡ௡

‫ ܩ‬ൌ ෍ ሺܽ௜௝ െ ௜ǡ௝ୀଵ

ሺ݊ ൅ ͳሻ ଶ ሻ ʹ

(6)

Value G is the sum of squared deviations of all the ranks from the mean value of the series.

݉‫ ܩ‬ൌ

݉ଶ ሺ݊ଷ െ ݊ሻ ൌ ܵ௠௔௫ ͳʹ

(7)

If tied ranks occur, the following formula is true:

݉‫ ܩ‬ൌ ܵ௠௔௫ െ ݉ܶ

(8)

Thus the value of the calculated Kendall's coefficient of concordance rating the significance of factors W=0,328. It has been suggested in literature [14] that the degree of consistency of Kendall's coefficient of concordance W should be described by the following hierarchy: - satisfactory: for the range between 0.20 to 0.40, - good: for the range between 0.41 to 0.60, - plus good: for the range between 0.61 to 0.80, - very good: for the range between 0.81 to 0.95, - ideal: for the range between 0.96 to 1.00. The above values allow formulating the conclusion that the degree of consistency of opinions among the experts participating in the survey is, according to the hierarchy proposed in [14], satisfactory. The assessment of significance of the coefficient of concordance has been done with the use of chi-square test with k-1 degrees of freedom. If there are tied ranks in the preference series, the value of chi-square is calculated on the basis of the statistics [3]: ߯௥ଶ ൌ

ܵ

(9)

ͳ ͳ ݉݊ሺ݊ ൅ ͳሻ െ ܶ ͳʹ ݊െͳ

where T:

ܶൌ

ͳ ଶ ଷ ܵ௠௔௫ െ ݉‫ ݉ ʹͳ ܩ‬ሺ݊ െ ݊ሻ ݉ሺ݊ଷ െ ݊ሻ ൌ ൌ െ‫ܩ‬ ݉ ݉ ͳʹ

(10)

The test of significance enables statistical assessment of the consistency of opinions among group of experts by formulating the following hypotheses: H0: Agreement among experts’ opinions is accidental. H1: Agreement of opinions is not accidental and the experts are competent. It has been read from the tables of real distribution S that for the number of degrees of freedom n= 15 and the level of significance α = 0.05, the critical value S is 24996. The data and results have been presented in Table 2.

257


Table 2. The results of statistical analysis investigating the agreement among experts’ opinions Number of factors

Number of experts

16

61

Kendall’s coefficient W

Calculated value of chi-square

Number of degrees of freedom

Assessment of consistency related to the significance of factors 0.328 300.120 15

Level of significance α

Critical value of chi-square

0.05

24.996

The test results have indicated that in both cases there are grounds for rejecting the zero hypothesis. It means that the agreement among experts’ opinions on the significance of factors is not accidental and the consistency of experts’ positions does exist. 4. Conclusions The tender procedure is the primary method of soliciting construction works contractors in Poland. The contractor’s decision to participate in a tender depends on many factors. The paper presents a group of factors which may affect contractors’ bidding decisions. Both the selection of factors and the rank of significance have been determined on the basis of opinions collected among Polish contractors. The assessment of agreement among the collected opinions has been done with statistical methods using the Kendall-Smith coefficient of concordance. The obtained results (the values of the coefficient, the test of significance) have confirmed the satisfactory level of agreement among experts participating in the survey and thus allow formulation of a general assessment of the factors and their ranks. It is worth noting, however, that the rating of factors influencing the decision to participate in a tender procedure is inevitably affected by experts’ subjectivity. The existing models supporting bidding decisions [8] take it into account and are based both on the assessment of a given factor’s influence on the decision in question and on the assessment of this factor in the context of each particular undertaking. References [1] I. Ahmad, I. Minkarah, Questionnaire survey on bidding in construction, Journal of Management in Engineering 4(3) (1988) 229-243. [2] A. S. Bageis, C. Fortune, Factors affecting the bid/no bid decision in the Saudi Arabian construction contractors, Construction Management and Economics 27 (2009) 53-71. [3] P. Cabała, Zastosowanie współczynnika konkordancji w pomiarze zgodności ocen ekspertów. Przegląd Statystyczny R. LVII – Zeszyt 2-3 (2010) 36-52. [4] D.K.H. Chua, D. Li, Key factors in bid reasoning model, Journal of Construction Engineering and Management 126 5 (2000) 349-57. [5] M. Gajzler, The idea of knowledge supplementation and explanation using neural networks to support decisions in construction engineering, Procedia Engineering, 57 (2011) 302 – 309. [6] M Gajzler, The support of building management in the aspect of technical maintenance, Procedia Engineering 54 (2013) 615-624. [7] M. Kozlovská, M. Spišáková, Modern methods of construction vs. construction waste , 11th International Multidisciplinary Scientific GeoConference, Conference Proceedings, 20-25 June Romania, 3 (2011) 483-490. [8] A. Leśniak, E. Plebankiewicz, Modeling the Decision-Making Process Concerning Participation in Construction Bidding, Journal of Management in Engineering 31(2) (2015) 04014032. [9] A. Leśniak, The simplified bidding decision model based on AHP method in ordering of investor’s supervision service, International Multidisciplinary Scientifif Geoconference, 14th GeoConference on Ecology, Economics, Education and Legislation, Mulitilateral Realtions and Funding Opportunities, Conference Proceedings, 17-26 June Romania, III (2014) 501-508. [10] A. Leśniak, E. Plebankiewicz, K. Zima, Design and build procurement system - contractor selection, Archives of Civil Engineering LVIII, 4 (2012) 463-476. [11] Public Procurement Biulletines, http://bzp1.portal.uzp.gov.pl/ [12] Report of the President of the Procurement Office on the Functioning of the Public Procurement System in 2013. Public Procurement Office, Warszawa 2014. [13] A. A. Shash, Factors considered in tendering decisions by top UK contractors, Construction Management and Economics 11 (1993) 111118. [14] A. Stabryła, Zarządzanie strategiczne w teorii i praktyce firmy, Wydawnictwo Naukowe PWN, Warszawa 2005 [15] Public Procurement Law -The Act of 29 January 2004 (PPL) Journal of Laws of 2013, No. 907, 984, 1047, 1473, of 2014 item 423, 768, 811, 915, 1146 and 1232) [16] M. Wanous, A.H. Boussabaine, J. Lewis, To bid or not to bid: a parametric solution, Construction Management and Economics 18 (2000) 457-466. [17] K. Zima, The concept of investment decision support model using fuzzy set theory,11th International Conference of Numerical Analysis and Applied Mathematics 2013: ICNAAM 2013. AIP Publishing ( 2013) 1307-1311.