A SMART METHODOLOGY FOR VALUE MANAGEMENT
Stuart D. GREEN, Department of Construction Management & Engineering, University of Reading, Whiteknights, PO Box 219, Reading, RG6 6AW,UK
Abstract
The paper considers the implementation of value management at the strategic interface between client organisations and capital projects. The ill-defined strategic problems which characterise the early stages of design will be contrasted with the well-defined problems which occur during the later stages. It will be further argued that the paradigm of traditional value engineering takes it for granted that design problems can be pre-determined and remain static over time. In contrast, the methodology of SMART value management recognises that it is not sufficient merely to 'achieve the required functions cost-effectively'. It is initially necessary to ensure that the key project stakeholders have developed a shared understanding of the strategic objectives. The methodology of SMART value management will be introduced as an approach which is both sound in its theory and practical in its implementation.
1.0 Introduction
The topic of value management is currently attracting considerable attention within the construction industry. The major clients of the 1990s are increasingly concerned with the achievement of value for money in their construction projects. In response to this demand, a growing number of construction companies and consultants are claiming to offer a value management service. However, 'value management' continues to mean different things to different people. There is considerable confusion regarding the distinction between value management and value engineering. Some commentators have even questioned the extent to which value management differs from the established procedures of cost management.
This paper will initially clarify the distinction between strategic value management and the narrower technical specialism of value engineering. To this end, the essential characteristics of traditional value engineering will be described and the underlying assumptions will be identified. On the basis of these assumptions, distinctive definitions will be developed both for the generic disciple of value management, and also for the occasional special case of value engineering. Finally, the improved approach offered by SMART value management will be described in detail.
2.0 Characteristics of value engineering
If value management is to be fully understood, it is initially necessary to appreciate the way in which it differs from value engineering. In its simplest format, value engineering can be defined as 'a systematic approach to providing the required functions at the lowest cost'. The methodology depends upon initially identifying what the required functions are, brainstorming techniques are then utilised to identify alternative solutions. The comparative merits of these solutions are then identified by the use of life-cycle costing techniques. The methodology of value engineering is well established and is often referred to as the 'job plan'. Whilst value engineering practitioners place great emphasis on adhering to the staged methodology of the job plan, it should be recognised that the underlying discipline is by no means unique. The five key stages are identical to the commonly accepted stages of creative problem solving:
Whilst the above approach can ensure that design decisions are better thought out, it is by no means distinctive to value engineering. If there is one characteristic which does make value engineering distinctive it is the emphasis given to function analysis. Indeed, almost every published definition is built around the word 'function'. Whilst the initial concept of function analysis was limited to the use of prompt questions such as 'what does it do?', it has since developed into a sophisticated pseudo-science known as 'function analysis systems technique' (FAST). Whilst the Society of American Value Engineers places considerable emphasis on the use of function analysis, the underlying theoretical basis is as yet underdeveloped. The available guidance on the use of FAST diagrams is notoriously opaque. There are also serious doubts regarding the extent to which function analysis is used in construction. Many attempts to use FAST diagrams produce an end result which is barely distinguishable from the traditional quantity surveyor's cost breakdown. Other practitioners merely use 'function analysis' as a label for an intensive group discussion which produces a list of problems relating to the current solution.
In the USA, the traditional vehicle for the implementation of value engineering is the 40-hour workshop. However, it is important to recognise that this is not the only approach. Indeed, within current UK and Australian practice, the 40-hour workshop is probably the exception rather than the rule. Enlightened practitioners in the USA are also increasingly moving away from the use of retrospective reviews by integrating 1-day VE workshops into the ongoing design process. It is surely preferable to focus attention onto 'getting it right first time', rather than subsequently correcting a poorly thought-out design.
There is a growing recognition that the maximum benefits of value engineering are to be gained during early design. However, in practice there is a paradox in that the earlier one tries to apply value engineering, the more difficult it becomes to define the nature of the problem. Many practitioners have compromised by advocating scheme design (RIBA Plan of Work Stage D) as the most convenient time of application. However, the decisions which really matter are invariably taken during the briefing and conceptual design phases.
It is important to recognise that the application of value engineering during the concept stage represents a markedly different problem from its use during detailed design. If this distinction is to be understood then it is initially necessary to challenge the underlying assumptions of the traditional approach to value engineering.
3.0 Underlying assumptions of value engineering
Whilst the rhetoric of value engineering is superficially appealing, it contains several important assumptions which are rarely acknowledged. The traditional literature on value engineering invariably assumes that design problems are both well-defined and static over time. Clients are further assumed to be unitary in nature and able to articulate objectives which are both consistent and transitive. Furthermore, it is invariably taken for granted that a 'system' possesses an objective underlying function which is waiting to be identified. It is also widely assumed that adequate decision support can be provided by the single-criterion cost models of building economics. Whilst these assumptions are by no means universal, they are strongly reflective of the currently dominant paradigm of value engineering. However, it is this same 'optimising' paradigm which has become increasingly discredited within the broader discipline of management science. The assumptions of value engineering seem especially naive when considered within the context of strategic decision-making by multi-faceted client organisations.
In contrast to the hard-systems tradition of value engineering, the intellectual origins of SMART value management lie firmly within the still developing field of soft operational research. The approach therefore represents a decisive break with the American tradition by rejecting both the optimising paradigm of hard-systems thinking and the associated terminology of function analysis. The broad theoretical framework for SMART value management is provided by the concept of group decision support (GDS), which can be defined as:
'....any designed process that supports a group of people seeking individually to make sense of, and collectively act in a situation in which they have power.'
The above definition presents an obvious analogy to the role of value management in aiding design decision-making in general, and the briefing process in particular. It should also be recognised that there is a significant difference between the provision of GDS and the narrower concept of decision support to an individual. GDS differs in that it places less emphasis on substantive data and more emphasis on consensus building and the decision-making process. Given that building design is invariably a group activity which includes both designers and client representatives, it is clearly GDS which is relevant. This is especially true for multi-faceted clients where different interest groups possess conflicting objectives. Indeed, once the intellectual baggage of optimisation is jettisoned, the concept of value management becomes almost synonymous with that of GDS.
4.0 Value engineering v value management: the distinction defined
From this point onwards it is convenient to assign particular meanings to the terms 'value engineering' and 'value management'. The following definitions have been developed by the Value Management Research Group at the University of Reading, UK. Whilst they are by no means universal, they are increasingly being accepted by leading client organisations and consultants.
Definition of Value Engineering
Value engineering is concerned with achieving a given function at minimum cost. It is based on the assumption that function is an objective characteristic which is waiting to be identified. Furthermore, it is assumed that all feasible design alternatives provide the same level of functional performance and can therefore be assessed on the basis of cost alone. Within this frame of reference, an increase in value can be directly related to a reduction in cost.
Definition of Value Management
Value management is concerned with defining what 'value' means to a client within a particular context. This is achieved by bringing the project stakeholders together and producing a clear statement of the project's objectives. Value for money can then be achieved by ensuring that design solutions evolve in accordance with the agreed objectives. In essence, value management is concerned with the 'what', rather than the 'how'.
Whilst the above definitions have proved to be influential in distinguishing the current best practice of value management from the cost-driven tradition of value engineering, it would be a mistake to perceive them as two different processes. Value engineering is best understood as a special case of the generic discipline of value management. When faced with well-defined technical problems, the value engineering approach is both valid and very powerful. However, when faced with the soft, messy and dynamic problems which often dominate during the early stages of building design it is the wider approach of value management which is necessary. There is little point in building cost-effectively if the ambiguity relating to objectives has not been resolved. A client may well build a facility in record time and at minimum cost. However, if it turns out to be a 'white elephant' it will never represent good value-for-money. In many cases the implementation of value engineering will only be meaningful following effective value management. For projects which are distinguished by uncertain objectives, it is value management which is relevant to the concept and feasibility stages, whereas value engineering will only become relevant to the later stages of the design process. It is also clear that the extent of client involvement which is needed for value management is considerably greater than that for value engineering. Indeed, it is arguably reasonable to expect value engineering to be carried out by the appointed designers and construction managers. This is where their expertise should lie. Many leading practitioners are increasingly trying to eliminate the need for retrospective value engineering studies by ensuring that the philosophy is integrated into the initial design process.
It can be argued that if value engineering is about building efficiently, then value management is concerned with improving the effectiveness of briefing. Issues raised often relate to the interpretation and communication of the client's strategic requirements. This is, of course, no easy matter. If the client is multi-faceted, as is often the case, then it is vital that the various interest groups are represented. The essential question which must by answered is: 'what does good value mean to the client ?'.Within the context of value engineering, the determination of 'maximum value' is likely to depend upon the use of economic analysis techniques such as life-cycle costing (LCC) or internal rate of return (IRR). However, the effective implementation of value management will invariably depend upon the consideration of multiple criteria, some of which will be objective and some of which will be subjective. If value management is to be implemented effectively, it must be based on an approach which is both rigorous and practical. Such an approach is provided by SMART value management which has been widely adopted by client organisations and consultants both within the UK and overseas.
5.0 SMART value management
SMART value management is based upon a technique known as the 'simple multi-attribute rating technique'. This approach has been specifically developed in order to combine the advantages of traditional value engineering with the soft-systems thinking which is required for effective value management. The SMART approach to value management usually consists of two one-day workshops. The first of these, known as VM1 should take place during the concept stage when the building of a new facility is first suggested as a possible solution to an identified 'problem'. The objectives of VM1 can be stated as follows:
Many construction projects invariably seek to satisfy more than one interest group, it is therefore vital that the decision-makers from each interest group attend the VM study. It should be stressed that all of these parties would have to be consulted in any case, the advantage of value management is that it provides a structured framework for discussion, thereby saving time. It also ensures that the problem is considered from all points of view before achieving a consensus view. Above all else, it ensures that the key stakeholders buy into the process and therefore have ownership over the resultant decision.The stages of VM1 shadow those of traditional value engineering and can be summarised as follows:
Stage 1: Information Each participant is invited to discuss their understanding of the problem together with their perception of the project objectives. The end product should be a list of agreed key objectives.
Stage 2: Structuring of Objectives The objectives listed in the previous stage are structured into a value hierarchy. The top of the tree is characterised by the overriding raison d'etre of the entire project. This is then progressively broken down into sub-objectives.
Stage 3: Speculation The generated objectives are then used as the stimuli for a brainstorming session; in what ways could these objectives be achieved? The success of this stage depends upon creative thinking. It is therefore important that ideas are not yet criticised or evaluated.
Stage 4: Evaluation The ideas produced in the previous stage are now evaluated. It is often possible to combine ideas to overcome any obvious shortcomings. Possible solutions should be subjected to careful consideration, however, they should not be rejected prematurely. In the first instance criticisms should be met with the response: 'OK, so how can we make it work?' The end result of this process should be a small number of alternative schemes which are considered worthy of further consideration.
Stage 5: Development It is likely each of the alternative schemes developed in the previous stage will need further development before their true merit can be assessed. It is important that the extent of the necessary follow-up work is agreed before the VM team disperse.
The second stage of value management takes place at the end of the outline proposal stage (RIBA Stage C) and is known as VM2. It is at this stage that the outline brief has been completed and the design team have produced a number of costed outline proposals. The objectives of VM2 can be listed as follows:
The nature of VM2 is therefore different from that of VM1, nevertheless it still focuses on the need for careful analysis. This is once again achieved by adhering to a staged methodology. However, it is likely that VM2 will involve more design professionals and less of the client's corporate personnel. There are seven key stages to VM2; the purpose of each stage can be summarised as follows:
Stage 1: Information The project is introduced by the project sponsor and discussion is directed towards the extent to which the value hierarchy established during VM 1 is still valid. The alternative design proposals will also be introduced by the design team leader.
Stage 2: Structuring of Objectives The identified project objectives are then structured into a value hierarchy. It is likely that this will be slightly different from that which was developed during VM1. It is also now necessary to simplify the hierarchy so that the lower order attributes can be used for the purposes of evaluating the alternative design proposals. The number of attributes can usually be reduced by eliminating those which do not directly influence the choice of outline design. It is usually preferable to omit capital cost until the end of the analysis.
Stage 3: Assignment of Importance Weights The importance weights are then assessed for each branch of the value hierarchy, the final weights are calculated by 'multiplying through the tree'. The weights should also be checked for consistency.
Stage 4: Evaluation Each outline design option is assessed on a scale of 0-100. A decision matrix is then used to obtain an aggregated utility rating for each design option.
Stage 5: Sensitivity Analysis The sensitivity of the evaluation to changes in the chosen importance weights and utility scores is tested by the use of sensitivity analysis.
Stage 6: Cost/Value Reconciliation The estimated capital cost of each proposal is then compared to the aggregated utility ratings and the decision is made as to which design option represents the greatest value for money.
Stage 7: Marginal Value Improvement The team then focuses attention onto the chosen option and identifies areas of concern. A brainstorming session is used to generate ideas as to how these concerns can be overcome. The workshop is concluded by a summary of what has been agreed together with the identification of any follow-up work.
6.0 The techniques of SMART value management
The purpose of this session is to explain and demonstrate the actual techniques which comprise SMART value management. On first sight, these techniques may appear to be overly intricate and complex. However, once the overall concept has been grasped, they are surprisingly simple. The secret to their successful use initially lies in understanding the underlying theory. It is then necessary to gain experience of their use in real situations.
It is convenient to consider firstly the structuring of objectives, assignment of importance weights, weighted evaluation and sensitivity analysis. This will then be followed by a separate consideration of brainstorming.
Structuring of Objectives
The ability to define and structure the objectives of a project is fundamental to the concept of value management. The techniques which are used to achieve this end are rooted in the field of decision analysis. They are also ideally suited to value management during the early stages of building design. However, it is important to recognise that the value hierarchy technique is concerned with establishing a shared perception of the design objectives amongst the stakeholders. It is not concerned with the identification of some sort of 'underlying truth'.
The first step is to structure the list of objectives which were agreed during the information stage into a value hierarchy. An example hierarchy for a proposed health centre is illustrated in Figure 1. The top of the hierarchy is characterised by the over-riding raison d'etre for the entire project. This is then progressively broken down into sub-objectives. Whilst the higher-order objective represents an end in itself, the lower order objectives are considered to be 'means-to-an-end'. It is important that the value tree is produced by group consensus and that each participant feels involved. Judgement is required with respect to how far the objectives should be subdivided.
During VM1, the object of the exercise is primarily one of definition and understanding. However, during VM2 the lower-order objectives are carried forward to the next stage. It is the 'twigs' of the value tree which provide the attributes against which the design options are evaluated. The required level of breakdown is therefore dictated by the need to compromise between the ease with which the attributes can be measured and the number of attributes with which the model can cope. An importance additional process at VM2 concerns the simplification of the value hierarchy. This can be achieved by asking the question "will this attribute influence our choice of outline design?". If the answer is "no" then the attribute concerned can be eliminated from the hierarchy. In some cases an attribute can be identified as a 'fundamental requirement' of all feasible design options. For example, objectives which relate to safety must be achieved in all alternative schemes. Therefore, it is often possible to prune these objectives from the value hierarchy. It is also necessary to ensure that two different attributes do not measure the same criterion.
It is often convenient simply to omit the objective which relates to capital cost from the hierarchy. This can then be re-introduced at a subsequent stage.
Assignment of Importance Weights
Having achieved a simplified value hierarchy during VM2, the next stage is to allocate an importance weight to each of the lower-order attributes. Once again, it is important that this process is performed on the basis of consensus. The SMART approach to weight elicitation is based upon the sum of the weights being equal to one for each group of attributes which derive from a single node. Each group is dealt with in turn and the final weights for the lower-order twigs are obtained by 'multiplying through the tree'. Figure 2 shows the previous value tree with the addition of importance weights. The relative weights for each group are determined by the ratio method. Attributes are initially listed in order of perceived importance and the least important is awarded an arbitrary weight of 10.
It is then necessary to allocate weights to the other attributes on the basis of their relative importance. The weights are then summed and each is normalised such that the total weight for the group adds up to 1. It is important that the facilitator regularly checks for the inevitable inconsistencies. The process is therefore likely to involve a good deal of iteration before the elicitated weights are fully consistent and the group is fully comfortable with the final result.
Weighted Evaluation
The allocation of importance weights to the lower-level attributes on the value hierarchy provides a rational basis for determining which outline design option provides the best value. The evaluation process is carried out by assessing each design option against each of the identified attributes. The assessment is performed by a process of scoring each design option against each attribute. Whilst the majority of scores will be allocated on a subjective basis, it is important that objective measures are used where possible. Subjective attributes are measured on an arbitrary scale of 0-100, where 0 represents the minimum acceptable standard and 100 represents the maximum which is achievable. For attributes which can be assessed objectively it is necessary to convert the measurements so that they are also represented on a 0-100 scale. Once all the scores have been produced they are 'weighted' by the appropriate importance weight. The weighted scores for each design option can then be summed in order to produce on overall score indicating the 'value' of each option (denoted Ui). The process is simplified by the use of the standard analysis form as shown in Figure 3.
If the issue of capital cost had been previously omitted, it is at this stage that it would be re-introduced. An additional step would calculate the ratio Ui/Ci where Ci is the estimated capital cost of the ith design option. It could then be argued that the highest value of Ui/Ci would dictate rational choice. However, if the budget limit is C*, then the highest value of Ui could be chosen provided that Ci is less than C*.
Sensitivity Analysis
The next stage of the process is to perform a sensitivity analysis. The purpose of this analysis is to test how sensitive the outcome of the rating process is to marginal changes in the key variables. Particular attention should be given to any importance weights or utility scores about which members of the group had expressed discomfort. It may well be thought necessary to adjust the structure of the value tree. The model would continue to be revised until the participants felt that it was reflective of their values for the project.In the final analysis, it is important to appreciate that the objective of the exercise is to provide a structured framework for thinking and communication. In no way does the use of value management techniques replace the expertise of the assembled team.
Brainstorming
It has long been recognised that people are capable of two distinctly different types of thinking: (i) creative thinking; (ii) analytical thinking. Analytical thinking is epitomised by the traditional thought process of progressing from one logical step to another. Creative thinking seeks to break out of this progression by making a 'conceptual step sideways'. Edward de Bono has compared analytical thinking to the process of 'digging the hole deeper' whilst creative thinking is seen as 'finding somewhere else to dig the hole'.
Creative thinking alone does not solve problems, it merely provides the alternatives from which to choose. Analytical thinking is then required to evaluate the practicality and the comparative merits of the identified solutions. Both types of thinking are necessary for solving open-ended problems; the merit of adhering to the structured approach of value management is that it ensures an equal emphasis is given to both. Research has suggested that without this disciplined approach decision-makers tend to revert to analytical thinking once the first workable solution has been identified. This state of mind is neatly defined as the single solution fixation.
Brainstorming is a disciplined technique which is deliberately aimed at overcoming the above defined single solution fixation. The approach aims to achieve the spontaneous generation of ideas in the absence of any criticism or evaluation. The quality of the ideas produced is not initially important; the objective is merely to produce a large number of ideas. Even wild or seemingly facetious ideas are of benefit in that they may act as 'intermediate-impossibles', i.e. as stepping stones to other ideas which would not otherwise have been identified.
The success of the brainstorming session is heavily dependent upon the ability of the facilitator to establish and maintain an appropriate climate. It is important that all participants appreciate the nature of the exercise and that criticism or evaluation is not allowed to take place. The facilitator must provide the momentum of the exercise and ensure that this is maintained. All ideas must be recorded on a flipchart as they are voiced; there is no such thing as a bad idea.Individuals whose natural tendency is to rely on analytical thinking may well feel uncomfortable in a brainstorming situation. It is the responsibility of the facilitator to ensure that any such negative feelings are overcome.
7.0 Summary
This paper has described the essential characteristics of traditional value engineering. It has been argued that the value engineering approach is only applicable to design problems which are well-structured and easily defined. The over-riding assumption is that 'function' exists independently of the conflicting and transient aspirations of the project stakeholders. It is taken for granted that function can be analyzed in the same dispassionate way in which one would manipulate a quadratic equation. Whilst value engineering may well provide a powerful cost reduction methodology during the later stages of the design process, it is rarely applicable to the 'soft' strategic problems of early design.
The SMART approach to value management is based on a way of thinking which is fundamentally different to that of value engineering. It has been specifically designed to deal with messy, dynamic and ill-defined problems. The approach is not so much concerned with 'problem-solving' as with establishing a common understanding of what the problem actually is. It is particularly important to dispel the notion that value can be optimised by the use of value management. In the final analysis, value for money has more to do with psychological comfort than it does with objective economics. It is also important to recognise that decision models cannot be used to automate the process of decision-making. The benefit of SMART value management is that it provides a framework around which the professional team can think and communicate. It also ensures that the decision-making process is explicit and rational. In the long-term, the use of SMART will ensure that practitioners reflect on their decisions, thereby learning from their experience.
BIBLIOGRAPHY
Connaughton, J. N. and Green, S. D., Value Management in Construction: A Client's Guide, CIRIA, London, 1996.
Green, S. D. 'A Metaphorical Analysis of Client Organisations and the Briefing Process' Construction Management and Economics, 14(2), 1996, pp 155-164.
Green, S. D. 'Sociological Paradigms and Building Procurement' in Proc. CIB W-92 Symposium, Procurement Systems: East meets West, (Ed. S. Rowlinson) Hong Kong, 1994, pp 89-97.
Green, S. D. 'A Paradigm Crisis in Value Management?' in Proc. 10th Annual ARCOM Conference, Loughbough University of Technology, 1994, pp 410-421.
Green, S. D. Beyond Value Engineering: SMART Value Management for Building Projects International Journal of Project Management, 12 (1), 1994, pp 49-56.
Green, S. D. 'A Reinterpretation of Value Management', in Proc. CIB W-65, Organisation and Management in Construction - the Way Forward (ed. T. M. Lewis), Trinidad, 1993, pp 437 - 445.
Green, S. D. Resolving Conflict in the Formulation of Building Design Objectives, in Construction Conflict: Management and Resolution (eds. P. Fenn and R. Gameson), Spon, London, 1992, pp 95-109.
Green, S. D. A SMART Methodology for Value Management, Occasional Paper No. 53 Chartered Institute of Building, 1992, 46+vi pgs.
Green, S. D. Value Engineering during Design: New Directions in Research, in Architectural Management (Ed. P. Nicholson), Spon, London, 1992, pp 256-267.
Green, S. D. and Moss, G. 'Value Management and Post-Occupancy Evaluation: Closing the Loop' in Proc. 5th European Value Management Conference, Brighton, 1994, pp 155-165.
Green, S. D. and Moss, G. Value for Money from SMART Management, Chartered Builder, 5, October 1993, pp 5-7.
Green, S. D. and Popper, P. A. Value Engineering: the Search for Unnecessary Cost, Occasional Paper No. 39 Chartered Institute of Building, 1990, 53+x pgs.
Green, S. D. and Simister, S. J. 'Theoretical Perspectives on the Development of a Meta-Methodology for Value Management', Proc. CIB W-92 Symposium, Procurement Systems: North meets South, (Ed. R. G. Taylor), Durban, South Africa, 1996, pp 168-179.