Finding The Best Procurement Strategy Construction Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Every time that a Public Authority has to deal with a bids of tenders, especially for a public work, it's difficult for it to decide which is the best company that will be able to do the works in the best possible way. Assuming that this inability of the client often leads him to adopt the same procurement route/methodology in each kind of project, the aim of this report is demonstrating or at least underlining that there are some researches about the possibility to discover a tender system that could be objective as much as possible and that could individuate the right procurement criteria for each single project. Even though there's a recognized existence of this problem, in general researches (Oyegoke, Kiiras, 2009) about it are still too few and a lot of studies are needed to investigate better this subject.

Anyway, after a short analysis of two methodologies found in the literature, it was possible to apply the easiest one to a real case set in Como, a city in the Northern Italy. The 'real' poor clear identification of the initial criteria can become more accurate if the Municipality is able to sub-divide the criteria in sub - criteria assigning to each of them a specific weight or a range in which the value must be found to be considered good.

In the end this sub-division of the criteria shows that the client needs to use different procurement routes/methodologies because it's demonstrated that a lot of details don't allow him to adopt the same behavior in front of different projects. In this way it will be assured the satisfaction of the client, of the companies and of all the people involved in the project.

1. Introduction

It is known that in general a procurement is a strategy to satisfy a client's development and/or operational needs with respect to the provision of constructed facilities for a discrete life cycle.

Most of the times, in the European Countries, this kind of strategy is used in public works with a specific goal to reach: to satisfy as much as possible the client's requires. This is the reason why there are some procurement's procedures that can be followed to achieve this purpose. Usually the most common public procedure, that is undoubtedly more complicated than the private one, is composed of a series of compulsory steps.

First of all the client has to choose a figure that is responsible of the entire procedure, that is usually an engineer; after that the 'Invitation for tenders' has to be stated. The 'Invitation for tenders' contains three important items: the definition of the project's objective, a list of tender documents that can change every time according to the client's needs, and requirements for the companies. Afterwards the companies decide to participate alone or in group like temporary or fixed societies of companies and one of these is chosen from the client through a specific procurement path. The three most common procurement paths used by the client to choose the company that will carry out the project are:

1. Opened procedure

2. Restricted procedure

3. Negoziated procedure.

The opened procedure provides for the participation of each company to the tendering process. In the restricted procedure each company can ask to participate but the offer can be sent only by some invited economic companies (above all when the criterion is that one of the most economically advantageous offer). In the last negotiated procedure the client consults the chosen economic operators and establishes with one or two of them the awarding conditions. (D.lgs 11 .09.08 n.152).

In the end, according to the previous requirements of the 'Invitation for tenders', the client awards a contract through two possible criteria: the lowest price or the most economically advantageous offer in order to enter into an agreement with the 'winner' company.

Some researches made in Norway (Lædre et al., 2006) established that most of the times there are some problems dealing with the procurement route that a client has to choose in order to find the most suitable company in doing works. Some research works taken from sources of literature, from support material of two software tools dealing with the selection of the procurement route and from documentation of public building and construction projects; showed that the proper procurement route will depends on the characteristics of each project. In fact from the same documentation it seems that public owners continue to select the same procurement route as they are in the habit of because usually they don't consider what procurement route suits each single project, and therefore they don't select the route according to a recommended practice. This report tries to support this theory demonstrating that if a different procurement route is used in public building according to a specific construction project, the final product will be a better result.

Through a correct procurement route it could be possible for the client not to risk to be unsatisfied about the product he obtains.

Following a recommended practice not only the client will be satisfied, but all people involved in a building competition will be interested in this topic; for instance the companies will be able to present their offers being aware of each specific require and criterion used by the client for the selection.

2. Support system procurement

In the literature there are some information on the procurement route strictly connected to the usefulness of a knowledge-based advisory system that can helps the client to optimize such decisions, admitting the multiplicity of project variables. Moreover this kind of topic is related to the awarding criterion of the most economically advantageous bid employing weights to aggregate the numerical scores assigned to each proposal with respect to different evaluation factors. In fact when a client has to choose a company in a procurement tender he should select different criteria according to the objective he wants to achieve. It's common that decisions to use any of the available approaches to procuring construction services are often subjective and because of that some clients use the same route for each project they have to deal with. Clients may also choose from a range of possible dispute resolution arrangements, payment terms and contractor selection methods. There is, however, an evident lack of consolidated knowledge about some of the specific merits of all such potential alternative routes.

For instance there are some alternatives that can be considered with respect to optimizing work packaging, team selection and risk distribution.

A recent research in Hong Kong has also confirmed the importance of a more enlightened assembly of appropriate procurement frameworks to suit particular project scenarios (Kumaraswamy, 2001), in particular thanks to these studies in China it was possible to identify additional potential criteria against which construction project performance may be evaluated. For this study, detailed information was collected from 22 practitioners, as well as from 18 building projects in Hong Kong.

As a first step these Chinese researchers have analyzed an hierarchical framework of the various procurement options developed in a preceding study; and have divided them in five principal sub-systems of a procurement system as `work packaging', `functional grouping', `payment modalities', `contract conditions' and `selection methodologies'. These were in turn structured into sub-sub-systems, for example: `valuation method', `payment timing', and `currency' (under the `payment modality' sub-system). Because of the function of a decision support system, that is assisting clients in making more informed procurement decisions, it was therefore decided to formulate a knowledge-based advisory system that could capture and consolidate experiences from previous projects.

The framework for the proposed decision support system was developed in stages, the first version was based on the initial concept of procurement-related contributors to project performance. However, after recent studies about the significance of non-procurement variables made through questionnaires and interviews, there were some changes as Fig. 1 shows.

Thanks to regression analysis the information was processed to obtain models that identified significant procurement and non-procurement-related variables influencing performance against both 'cost' and 'time' targets.

From the preliminary interviews made in Hong Kong 38 performance criteria were obtained, but among them only 11 performance criteria (PC) were selected.

(PC1) Lower capital cost

(PC2) Lower life cycle costs

(PC3) Cost certainty

(PC4) Shorter pre-construction duration

(PC5) Shorter construction duration

(PC6) Time certainty

(PC7) Higher quality

(PC8) Effective and efficient communication

(PC9) Effective and efficient decision making

(PC10) Dispute minimization

(PC11) Overall client satisfaction (also including other aspects).

Fig. 1 points to the structuring of three knowledge bases at the core of the proposed model. These are designed to capture and incorporate any relationship linking project performance levels (as measured against the above criteria) against each of the following three sets of independent variables: (1) project-specific internal conditions (e.g. project complexities); (2) procurement options; and (3) external conditions (e.g. skilled labour availability in the project location). In this model there is also a set of relationships, that could both constrain and guide the procurement selection process and that arise from the need to choose `compatible procurement options'.

Fig. 1 - Proposed model of decision support system for optimizing procurement protocols and parallel managerial sub-systems.

Source - Kumaraswamy M.M. & Dissanayaka S.M., 2001

In this model there are User inputs ( I1 - Expert system ) that are used to model the project profile, for comparisons against the assembled knowledge-bases. In an intermediate stage 'Human' User inputs (I2) are required to check on compatibilities pending further knowledge-base development (at F). In the end User inputs (I3) recheck overall compatibilities, special conditions and any unforeseen potential side-effects of up to three potentially suitable procurement systems. The decision support system is developed to suggest appropriate parallel management sub-systems (e.g. for planning/scheduling, monitoring, quality and safety) that would be compatible if not synergistic with the recommended procurement system. These would also help to address any significant non-procurement variables that may affect performance, despite appropriate procurement decisions.

In Fig.2 the principal interactions within the modeled scenario are shown; the main variable sets are taken as internal conditions (I), external conditions (E), procurement options (O) and performance criteria (P). For each of the four variable sets, databases are necessary. On the other hand knowledge-bases are useful to model the relationships between them, such as at the interfaces IO, IP, EO, EP and OP. All these relationships are modelled in the knowledge-base modules 1(A), 1 (B), 3(A) and 3(B) in the knowledge-base 2 as it is showed in the Fig.1. IOP, EOP are three-way relationships modelled in knowledge-base 1 and 3. IEPO on the other hand, represents the fully integrated interface which would lead to suggested options (OS) for a specific project, based on the prevalent project-specific internal conditions (IS), external conditions (ES) and targeted performance profile (PS).

Fig. 2 - Schematic representation of interacting sets of variables and the necessary databases and knowledge-bases

Source - Kumaraswamy M.M. & Dissanayaka S.M., 2001

In 1998 a pilot survey was conducted focusing on building projects in Hong Kong to develop and test modules of the proposed system. Eleven performance criteria were selected among 38 initially assembled for the performance criteria (P) database. Both the preliminary interviews and studies fed into the selection of variables to be incorporated in the I, E and O database. The four sets of variables were incorporated in two data-sheet type questionnaires (Kumaraswamy, 2001).

One data-sheet questionnaire of 13 pages sought the experience-based perceptions of expert/more experienced building professionals on the influences of each of the selected procurement option-related (O) and non-procurement-related (I {internal} and E {external}) variables - on each of the performance criteria (P variables). This questionnaire is divided into three parts:

Part I, relating to the respondent and his/her organization;

Part II, relating to the OP relationships and

Part III, to the IP and EP relationships.

A final page invited overall impressions of potential impacts of each `factor category' of the `Internal Conditions', as well as of the `External Conditions': on the performance level against each of the 11 performance criteria in turn.

The other data-sheet `questionnaire' solicited information about specific projects - i.e. on the impacts of the same (O, I and E) variables on the performance levels (P) attained in those projects (against each of the eleven performance criteria).

The complexity of the data-sheet questionnaires was necessitated by the detailed data, that admit the potential interactions amongst these variables. However, it is also important to improve the reliability of the responses and to reduce the apparently high demands on the respondents.

In this case fifty experienced building professionals were approached. A total of 40 detailed responses were obtained after follow-up by the Research Assistants. The `experienced-based' data-sheets were filled in by practitioners working for the following organizations types: clients, client/consultants (clients who had in-house consultancy teams) consultants and contractors. This was considered a reasonably representative distribution to enable a preliminary assessment of building construction industry perceptions on these matters in Hong Kong.

2.1 Results

The perceived impacts of variables (both procurement and non-procurement-related factors) on project performance were at first assessed by respondents against a seven-point scale (from one to seven) in the data-sheet questionnaire. Afterward these scores were changed to a scale from -3 through to +3 for convenient analysis. In this scale -3 represented 'extreme negative impact', 0 represented 'no impact' and +3 represented 'extreme positive impact'.

From the literature it's possible to understand how this system works through a series of exercises. One of these is based on the perceived 'average impact value' of each procurement variable (option O) upon each of the 11 performance criteria (P).

For the ith procurement option the following formula has to be considered:

where AIVij is the average impact value of the ith procurement option on the jth performance criterion; Xijk is the impact value of the ith procurement option on the jth performance criterion as perceived by the kth respondent; and N is the number of respondents.

The indicative average perceived impact value of each procurement option (in turn) on overall performance was then approximately assessed as follows:

For the ith procurement option,

where (AIVi )op is the indicative average impact value of the ith procurement option on overall performance; AIVij is the average impact value of the ith procurement option on the jth performance criterion; and M is the number of performance criteria considered.

The (AIVi )op parameter by itself may be considered unrealistic but in reality all chosen performance criteria are considered in that value to be of equal importance in a given project. Therefore, the potential overall impact values obtained in this sub-exercise are merely indicative, and should be referenced only where no particular project is under consideration.

In the Fig.3 the table shown contains an example of a module from the pilot knowledge-base developed following these two steps analyzed. For example, the highest (AIVi )op value of 0.97 is obtained for FG3 (Design and Build) indicating that this sub-system is perceived to have a higher overall positive impact on project performance levels against the M chosen criteria when you consider that all criteria are equally important in the first instance. In conclusion of these studies procurement choices must be approached from a more holistic perspective, rather than merely optimizing OP (procurement option-performance) relationships by themselves (Kumaraswamy, 2001). It was evidently necessary to simultaneously incorporate relationships with the non-procurement variables such as IP (internal condition-performance) and EP (external conditions-performance), as also initially postulated in the model illustrated in Figs. 1 and 2. These relationships themselves interact with the OP relationships. On the other hand it is possible that not much effort is devoted to a structured pre-procurement study of priorities in most projects; which if done may lead to a choice of more performance-oriented procurement options.

Fig.3 - Extract from knowledge-base for 'Functional Grouping' procurement sub-system

Source - Kumaraswamy M.M. & Dissanayaka S.M., 2001

3. Post-objective determination of weights

Now it's useful to underline that more often the common tendency of a client to 'pursue' an objective is that one to reach the maximum attainable public value, commonly called value for money. The lowest price criterion for contract awarding is based on the notion that, if a bid complies with the minimum requirements of preset tender specification characteristics, the only evaluation factor to consider is price. As a result, proposals which may provide enhanced characteristics possibly at slightly higher price, don't receive any advantage in the evaluation process. In this way, important evaluation factors such as technical merits, quality, experience, extent and length of guarantees, maintenance cost, after sale service and life-cycle cost, are ignored.

On the other hand the criterion of the most economically advantageous tender considers non-price factors together with price in the evaluation of proposals but it's fundamental that evaluation factors should be publicly announced in advance of the tender. In this case every proposal receives a score for each non-price evaluation factor. The awarding decision is made by selecting the proposal which maximizes the ratio of the aggregated score divided by the price: this procedure provides a more complete comparisons among the proposals.

In a specific research made by Lorentziadis (2008) it's reported that weights, fixed and announced in advance at the Request of Tender, should be obtained by subjective judgment. But, more objective and fixed are these weights, more limited corruption there will be in the evaluation of the proposals. Therefore it's possible to create another kind of evaluation system in which too much emphasis is placed on particular evaluation factors, thus favoring bidders that score high in the corresponding factors.

From the literature another evaluation system is available, easier than the previous one, that combines pre-subjective input and post-objective determination of the weights.

Before the submission of the offers, intervals within which the weights should lie are subjectively set by the issuer of the tender call. This is a pre-subjective input to determine the weights, which becomes restrictive if the width of the interval is small and is relaxed for broader intervals. After that the proposals are unsealed and become known; the proposed methods objectively determine the exact value of the evaluation weights, based on information stated in the bids. This post-objective approach of weight determination takes into account, as a constraint, the pre-subjective input.

The weight distribution may be derived as an aggregate measure, like an average, of the least favorable distribution, or, alternatively, the most favorable distribution, that each bidder may face. Afterwards there is a variety of post-objective methods of weight determination analytically developed.

Usually in a competitive public tender there are N bidders who submit sealed proposals with prices pi, i = 1,. . . ,N. The price is, in general, a numerical monetary measure, which may, for example, be the direct building cost or the annual cost of operation. The evaluation and awarding criterion is the most economically advantageous tender. There are M evaluation factors which are publicly announced in advance of the bid. After the opening of the bid the performance of every proposal is rated for each evaluation factor and receives a set of scores denoted by Xij, j = 1, . . . ,M; where i = 1, . . . ,N. The values of Xij can be expressed within a numerical scale from 0 to 100, or, in general, within the interval [0, 1]. The numerical attributes are, then, aggregated by a set of weights wj, j = 1,. . . ,M, producing for each proposal the aggregate measure.

The most economically advantageous proposal is determined by minimizing the ratio of the price pi over the aggregate measure Ai. Typically the weights are fixed but in this case the literature proposes a method for the determination of weights like a good alternative.

In general it is assumed that the tender call sets a wide requirement for the magnitude of the weights and, in particular, intervals [Wj;W'j] are defined within which the weight wj should lie. The definition of the intervals [Wj;W'j] before the submission of the bids introduces a pre-subjective component in the determination of weights and should reflect the importance of each evaluation factor to the issuer of the tender call. After that two sources of subjectivity are introduced in the tender process: the first one is the choice of the particular post-objective weight determination method, and, the second one is the definition of the weight intervals [Wj;W'j]. Both these subjective elements of the evaluation process are publicly announced in advance and apply uniformly to all bidders.

In addition, it will be assumed that:

since otherwise the sum of the weights wj exceeds 1.

The vector of least favorable weights v(i) =( v1(i) ,…., vM(i) ) of bidder i is defined as the solution of the following minimization problem:

subject to the constraints:

The weight vector v(i) corresponds to the least favorable weights by which the proposal of bidder i could be evaluated. The Least Favorable (LF) evaluation method, awards the contract to the proposal with the minimum value of hi. Under this procedure, each bidder is evaluated with a different set of weights, which may raise some concern about the objectivity and equal treatment of the bid (Lorentziadis, 2008).

4. 'The empirical data'

The objective of this section is studying a possible way to award a tender in the reality, according to the literature just quoted. It will be used like a mere example an "invitation for tender" made by a Public Contracting Authority, the Municipality of Como, a city in the North of Italy.

In this case of study it is assumed that the Municipality of Como is interested in a requalification of a public area through the construction of a building used like a library and a surrounded green area like a park. In reality what the bids for tenders says about the awarding conditions is the typology of criterion chosen by the Municipality: 'the most economically advantageous offer'. In the bids of tenders the Municipality of Como looks for the values (stated in points) of some indicators that it usually used to apply in each kind of project it has to deal with.

Therefore the aim of this section of the report, according to the literature, is that one of applying one between the two illustrated methods. In this specific case it will be used the second one regarding the weights' attribution.

Following the literature, first of all the Public Contracting Authority (the Municipality) must specify that the contract award criterion is the most economically advantageous tender. After that, to ensure respect of the principle of equal treatment in the procurement process and secure the necessary transparency in order to make every tenderer aware of the criteria and methods to be applied for identifying the most economically advantageous tender; the weighting scheme of the award criteria must be included in the tender documents.

To attribute the weights and intervals of weights (according to the literature's method), it's fundamental to individuate the parameters that have to be considered to award the tender. This is a main point because to obtain a good result it is convenient to divide and sub-divided as much as possible the criteria. For instance, to redevelop this public area the 'quality' of the final product can be judged more easily by determining individual criteria, the fulfillment of which shall lead to a safer understanding of the level of quality of the product. In this case it will be useful to break down the criterion of quality to individual criteria: completeness, availability, reliability, flexibility, timeliness, security, accuracy, usability.

These sub-criteria will facilitate the task of the body responsible for the evaluation of tenders, on the one hand, and will lead to safer choice of the best service from those offered or of the best product in terms of quality, on the other (Fig. 4).

The 'technical value' of the offered solution could be replaced with criteria referring to the methodology and support tools that the tenderer intends to apply in the implementation of the contract.

The 'environmental characteristics' can be judged in terms of performance or functional requirements and are appropriate for determining the characteristics that form the contract scope.

In this case of a public service contract the efficiency can be checked using various criteria referring to the breakdown of the project into individual work packages and deliverables, or to the definition of the contents of the individual deliverables, as well as to the proposed method for organizing and coordinating the key experts. In the case of public supply contracts, efficiency may be judged based on quantitatively identifiable data, such as the number of printouts of a printer under evaluation or the number of a computer's network transactions over a specified period of time.

'After-sales service and technical assistance' may be checked using various criteria referring to the service/support methodology that the tenderer intends to apply, the geographical coverage of the network of the tenderer, or to the method used to organize the technical service and assistance procedure.

'Understanding the requirements of the contract' is also a decisive criterion, especially in the case of complicated contracts for complex projects, where successful implementation of the project depends mainly on the candidate's ability to understand the objectives and specifies ways in which it's easy to address potential risks.

The 'implementation schedule' presented in the tender (with the indication of the critical milestones) is also a criterion whose effective use ensures selection of a competent contractor who has understood the requirements of the Contracting Authority, especially in cases where the interim critical milestones of the contract scope under implementation are not fully and precisely in full detail in the tender documents (Republic of Cyprus, 2007).

In this way, after the accurate subdivision of the main criteria in more detailed sub-criteria, according to the importance that the client wants to associate to these parameters, it will be easier for him to establish the weight for each single criterion and above all the range of weights [ Wj ; W'j ] in which the value of the criterion has to be individuated to be accepted.

As it is showed in the literature, the vector of least favorable weights v(i) =( v1(i) ,…., vM(i) ) of bidder i is defined as the solution of the minimization problem:

The value of hi that has the best result among all the companies involved in the bids of tenders is that one of the winner company.

5. Conclusions

Analyzing the case of study just described it's clear to understand that a client must take into account that it's impossible for him to chose a company with the same procurement route in each project he has to deal with. This theory is shown above all by the possibility to subdivide all the criteria in sub-criteria; this means that there are a lot of details that have to be considered when a company's choice has to be done.

This report has the task to demonstrate that there are some researches dealing with this topic but not enough to find a possible objective method that can individuate with an easy and an immediate system the best company for a specific project. But, the reality has pointed out that a lot of "Public Authorities" award a tender without thinking to a proper procurement methodology of choice. The two methods showed in the second section of the report, taken by the literature, are only two examples of what researchers are doing in order to achieve a more clear objective methodology even though there are some parts of them still subjective. Anyway, in the end, it could be stated that, even if there isn't a specific methodology to identify the correct procurement route to adopt, this study has been very useful to underline that the first fundamental step versus the objectivity of the methodology is the subdivision of the criteria in fact more criteria are considered, more accurate will be the attribution of the weights to all the parameters involved. To conclude, using a correct procurement route, on one side the participating companies are more aware of the client's needs bringing, in the same time, more satisfaction to the client.