Open individualized approach
Former för ansökan Beredning Bidragsgivande
Claes Hedberg
2011-09-01 22:31, reviderat 2011-09-01 22:32
The success of SSF depends on the success of individual projects. Success of a project is commonly determined after its completion - at best. They may actually to a high degree of accuracy be assessed during the project.
The role of SSF is suggested to be evaluator and rewarder of the quality of the projects rather than the proposals.
It is possible to approach the reasoning above, by putting more emphasis on two aspects:
1) To make yearly evaluations of the results with the option to halt financing.
2) To introduce a competition aspect.
This would free resources so that more projects may be started, and thereby more tracks would be followed.
The strategy for project approval may be distributed by letting the proposal decisions be delegated to individuals of the research world.
Strategy can be project oriented.
For SSF, it should be, for two reasons:
A. The success of new ideas and changes are dependent on getting particular details right . These can only be found in individual projects, which could or could not be under a strategic umbrella - for the success it really does not matter.
B. The object of SSF is to support new discoveries, new products and new companies. Again, these can never be found in strategies, only in the specifics of one or a few researchers or inventors.
It is therefore proposed that there are no prior directions of research, apart from the wide subject areas of engineering, natural sciences, and medicine.
Proposed tentative guidelines of system
Actor = group doing research in a project
A project may have more than one actor, where they are working on the same problem statement.
Let a system be organized, which lets any research group make short applications for what they want to do.
* free topics within engineering, natural sciences, and medicine.
* free actors (universities, institutes, companies, .. )
* yearly evaluations of projects. Half of the participating actors should be terminated each year
* no cooperation between academia and companies is needed, but welcome when it increases work quality.
* the efforts should be put into the evaluations, and not so much into the project applications. This is to reward quality of work, and not of applications.
* actor competition. Many projects should in their initial phases have at least two competing groups working in parallel on the same task, making even clearer the focus on performance.
* the project evaluations will be done by different (from the full time evaluators) people, preferably from abroad. These will distinguish and rank the projects. These will be done anonymously, so that the evaluators will not know who did which report. This is how quality work will be rewarded.
* to prevent inbreeding, it is proposed that a certain percentage of funding should be given to actor groups which have not earlier, eg within the prior 10 years, obtained support.
* each project can get support for at most 4 years.
Outline of work for evaluators
Let's fix the support from SSF to each approved actor to be 1.5 million SEK per year.
This means that there can be 600/1.5 = 400 actors supported each year.
An evaluator picks the most promising of the project proposals, and evaluates running projects.
An evaluation of a new project should take one day. An evaluation of a running project should take 10 days. Half of the existing project should be continued each year, with the rest being terminated.
Each evaluator can then evaluate 14 projects and 70 new ones. This finances every year 10 % of the new ones, and half of the old ones. Each evaluator chandles the same amount: 7*1.5 + 7*1.5 = 21 millions per year, which leads to 600/21 = 30 full time evaluators. This is based on 30*70 = 2100 project applications per year. (How many are processed today ? How many can be expected in the new system ?) The percentage of approval rises if less applications.
The estimation of administration cost for this evaluation system is on the order of totally 60 million = 10% of the distributed budget.
The evaluators should come from diverse backgrounds - why not similar to the suggested groups in the SSF strategy process ? They should work as evaluators for two years.
Claes Hedberg, professor
Blekinge Institute of Technology, Research director of Structural Analyses
Inga kommentarer inlagda.
Fasen är stängd, inte möjligt att kommentera.