Service Communautaire d'Information sur la Recherche et le Développement - CORDIS


SAPIENTIA Résumé de rapport

Project ID: ENK6-CT-2002-00615
Financé au titre de: FP5-EESD
Pays: Greece

Construction of the ISPA tool - Integrated R&D policy exploration using ISPA

The PROMETHEUS stochastic model has provided the expected impacts and variance-covariance matrices of the impacts that were used in constructing ISPA, the policy integration tool used to perform Integrated R&D Policy Exploration. In SAPIENTIA ICCS-NTUA has undertaken a new specification for ISPA in which the expected impact on a given objective is a non-linear function of the R&D allocated to this technology. No numerical problems arise, and seemingly convexity is retained. However, a considerable increase in complexity is evident. In general, results are defendable and exploration has proceeded in a similar fashion as with the earlier version of ISPA, but they are generally highly sensitive to budget size. The integrated exploration procedure adopted using the ISPA tool includes the following steps: construction of a feasible set by exploiting synergies among the objectives,improvement of the solution obtained by consolidating synergies and relaxing bounds, and, finally, improving the solution by sacrificing the probability (or the threshold) requirements for some objectives. Guidance in the relaxation and the sacrificing has been provided by the shadow costs. A Mixed Integer Programming (MIP) specification of the ISPA model has also been tested, which provides complete flexibility on the joint distributions of the impacts and expands the possibilities for adopting different risk averse stances. However, the MIP version has some disadvantages, such as computational difficulties, difficulties in introducing non-linearities, and, limited possibilities for using shadow costs for guiding the exploration. The exploration procedure adopted using the MIP version of ISPA includes the following steps: setting ambitious expectation targets for all objectives, then giving emphasis on one objective at a time by seeking higher expectations and lower risks, and, consolidating gains both in terms of expectations and risks by exploiting synergies between the objectives. The solution obtained allocated 20-26% of the budget to non-conventional vehicles, 22-26% to nuclear, 25-27% to renewables, and 17-21% to clean coal technologies. Fuel cells share is around 10% when the market impact objective is emphasized, but drops to 3.5% when emphasis is placed on other targets. The solution obtained is diversified in terms of technologies included and it is fairly stable despite the shifts in the emphasis on impacts.

Informations connexes


Pantelis CAPROS, (Scientific Responsible)
Tél.: +30-2107723629
Fax: +30-2107723630