Structuring resource allocation decisions: A framework for building multi-criteria portfolio models with area-grouped options
Introduction
Allocating scarce resources to projects or services within organisations involves the careful balancing of potential benefits against costs, and an appraisal of the risks of realising such benefits. Although this is a common task for managers working in a business planning role, it is one that is far from being straightforward. Recently, Phillips and Bana e Costa, 2007, Kleinmuntz, 2007 cited several challenges faced by managers in charge of allocating resources, which can be summarised as follows: (1) there is usually a large number of potential projects and a limited amount of resources; (2) benefits are typically characterised by multiple and (often) conflicting objectives; (3) no manager has a complete understanding of all the consequences of every project – instead such information is spread across different organisational layers and departments; (4) the allocation of resources to organisational units considered individually will not necessarily result in a total allocation that is collectively efficient; and, finally, (5) if the resource allocation process is not properly managed, it may lead managers to invest in projects that are not necessarily aligned with the organisation’s strategic objectives.
Multi-Criteria Decision Analysis (MCDA) is an approach that has been claimed as an effective way of assisting managers in tackling the challenges described above and has been extensively employed in practice (Phillips, 2007, Phillips and Bana e Costa, 2007). Several MCDA-based modelling approaches for resource allocation exist (for an overview see Archer and Ghasemzadeh, 1999, Kleinmuntz, 2007), most of which consider a single set of projects which are evaluated by multiple criteria (e.g. Golabi et al., 1981, Liesiö et al., 2007, Stummer and Heidenberger, 2003). Some of these approaches, however, also allow for the possibility of grouping projects into organisational areas (e.g. departments, regional offices, etc.), and the focus of this paper is on this type of modelling method.
Despite the growing attention to the structuring phase of MCDA interventions (Bana e Costa et al., 2002, Bana e Costa et al., 1999, Barcus and Montibeller, 2008, Belton and Stewart, 2002, Brugha, 1998, Brugha, 2004, French et al., 1998, Keeney, 1992, Watson and Buede, 1987), there is still little indication in the operational research and decision sciences literatures on how to structure these models in practice. The general purpose of this paper is thus to address this apparent gap, and propose different approaches to structuring multi-criteria portfolio models with area-grouped options. We illustrate the use of these approaches with a couple of real-world case studies, drawn from action-research projects conducted by the authors. Reflecting on the experience, we then suggest a general framework for guiding the structuring of this type of model.
The rest of the paper is structured as follows. The next section describes the problem of selecting a portfolio using multi-criteria analysis. The following section discusses briefly the nature of problem structuring in organisational decision making, characterising it as a complex social process. In the subsequent section we discuss how multi-criteria portfolio models with area-grouped options can be structured in different ways, by presenting alternative approaches to structure decision areas, options and evaluation criteria. Two action-research projects, in which particular combinations of these different structuring approaches were employed, are then presented and discussed. In the subsequent section, we suggest a general framework for the structuring of these models, derived from the experience in applying these approaches in practice. The paper concludes discussing some issues for the research and practice for this type of modelling.
Section snippets
Multi-criteria portfolio models with area-grouped options
The use of MCDA for portfolio analysis, where projects are grouped into areas, involves a different problematique (Roy, 1996) than the one traditionally dealt with in MCDA applications. The set of options is distributed in K areas, which may represent organisational units such as departments, geographical offices, or other relevant organisational groupings. As shown in Fig. 1 below, each kth area has nk options, which are denoted by oki (i = 1, 2, …,nk). Some of these areas may have cumulative
Problem structuring in organisational decision making
It has been widely acknowledged that problem structuring (also known as ‘problem formulation’) is a critical activity in organisational decision making (e.g. Dutton et al., 1983, Eden, 1986, Franco et al., 2007, Lyles, 1981, Lyles and Mitroff, 1980, Mintzberg et al., 1976, Nutt, 1992). A full discussion of problem structuring is beyond the scope of this paper, but for our purposes it would be useful to conceptualise it at least at three levels of analysis: individual, group and organizational.
Structuring multi-criteria portfolio models
As already stated in the introduction, practitioners and researchers of MCDA have increasingly recognised the importance of problem structuring in real-world interventions. Two activities are required when structuring a standard MCDA model: developing a hierarchy of criteria/objectives and defining/identifying a set of options.
As we have seen above, however, multi-criteria portfolio models with area-grouped options have a more complex structure of options. There is not, to the extent that we
The decision problem
The severity of the problem of teenage pregnancy in the UK was highlighted in a report by the Social Exclusion Unit (SEU, 1999) which set demanding improvements to be achieved by 2010. Up to 2005, the teenage pregnancy rate in the UK (42.3 conceptions per 1000 girls2) had remained similar to that in the 1970s, while in most of Western Europe it had halved. This problem exhibits all the typical characteristics of a complex
The decision problem
Our client in this case was the Library Learning and Culture division of a large regional county located in the UK Midlands, hereafter the LLC. The LLC is divided into six subdivisions (libraries, museum service, county record office, adult and community learning, heritage education service and county arts service), and its aims are to inspire learning and imagination for people of all ages living in the region.
In the past, LLC used to receive a large level of funding to invest in its
A framework for structuring multi-criteria portfolio models
The two case studies just described showed two different ways of structuring the areas & options and the criteria. Reflecting on the case studies, it is possible to draw a more generic framework for structuring multi-criteria portfolio models where the two approaches for structuring options & areas and the two approaches for structuring criteria can then be combined, depending on the type of decision problem that the facilitators/analysts are supporting and what they want to achieve.
This
Conclusion and directions for further research
Supporting the allocation of resources with multiple criteria portfolio analysis models has been increasingly employed in practice. These models help decision makers to identify the combinations of options, from each organisational unit, that generate greater value for the organisation – i.e. a set of efficient portfolios. Structuring this type of model is not a trivial task, and as far as we are aware, no previous studies have attempted to understand systematically this process. This paper was
Acknowledgement
We would like to thank Alec Morton and Larry Phillips for the useful suggestions of references on the topic and insights from their experience. We are also grateful to three anonymous referees for their detailed feedback and useful suggestions on earlier versions of this paper.
References (49)
- et al.
An integrated framework for project portfolio selection
International Journal of Project Management
(1999) - et al.
Decision support systems in action: integrated application in a multi-criteria aid process
European Journal of Operational Research
(1999) - et al.
Facilitating Bid evaluation in public calls for tenders: a socio-technical approach
OMEGA
(2002) - et al.
Supporting the allocation of software development work in distributed teams with multi-criteria decision analysis
OMEGA
(2008) The structure of development decision making
European Journal of Operational Research
(1998)Cognitive mapping: a review
European Journal of Operational Research
(1988)Analyzing cognitive maps to help structure issues or problems
European Journal of Operational Research
(2004)- et al.
Preference programming for robust portfolio modeling and project selection
European Journal of Operational Research
(2007) - et al.
Behavioral and procedural consequences of structural variation in value trees
European Journal of Operational Research
(2001) - et al.
Group support systems for multi-organizational collaboration: reflections on process and content
Group Decision and Negotiation
(2005)