The selection of joint projects by a consortium: Cost sharing mechanisms
- PDF / 161,052 Bytes
- 8 Pages / 595 x 842 pts (A4) Page_size
- 105 Downloads / 152 Views
#1999 Operational Research Society Ltd. All rights reserved. 0160-5682/99 $15.00 http://www.stockton-press.co.uk/jors
The selection of joint projects by a consortium: Cost sharing mechanisms JA Aloysius1* and EC Rosenthal2 1
University of Arkansas, USA and 2Temple University, USA
This paper considers the problem of project selection and cost allocation for a partly decentralised organisation such as a research consortium, whose members have conflicting preferences and limited budgets. Three normative properties that project selection and cost sharing mechanisms which should satisfy are proposed. We introduce a class of efficient mechanisms called willingness to pay that satisfies the properties and solves the interdependent selection and allocation mechanisms through mathematical programming. These mathematical programming procedures are shown first, to improve upon existing cost sharing plans used in practice, and second, to be undominated by any other selection and allocation mechanism that satisfies the properties. However, in the case of private information the procedures are not incentive compatible. For this case, we provide an incentive compatible, though inefficient, mechanism, and prove that no efficient mechanism can exist for this class of problems. Keywords: project selection; cost allocation; capital budgeting; game theory; integer programming
Introduction In today’s increasingly competitive environments, companies face ever more difficult problems in deciding what research and development (R&D) projects to fund. Such problems are even more complex for partly decentralised multi-divisional organisations such as research consortia, since the various members, with disparate objectives and needs, are in conflict with one another even as they pool their resources toward joint project development. Research consortia have become integral to several high-technology industries since the mid-eighties.1 The general model we assume is one where each of the proposed projects has a given cost, and each of the members reports projected benefits from obtaining use of the resulting technologies. We will analyse the optimal allocation of the members’ limited research budgets toward the funding of the most profitable projects to the consortium as a whole. We study project selection and cost allocation schemes, with two aims: (1) to select the optimal subset of projects to fund; and (2) to fairly allocate the cost of the projects across the members. These goals are interrelated, because the cost allocation method used has implications for the selection process, and therefore for efficiency. The applicability of the study goes beyond research consortia, to any multi-divisional organisation where the divisions have different requirements but the organisation *Correspondence: Dr. JA Aloysius, Department of Computer Information Sciences and Quantitative Analysis, 204 Business Administration, University of Arkansas, Fayetteville, AR 72701, USA.
as a whole has a unified objective. Actions optimal for one unit may be suboptimal f
Data Loading...
