Multicriteria analysis for the selection of priorities in the Brazilian program of technological prospection--Prospectar.

SUMMARY

This article describes a methodology of organization for the data obtained by Prospectar, aimed at the selection of technological priority topics and a strategic use of the information collected by the Program. The challenge of defining priorities on the basis of an exhaustive prospective exercise, like Prospectar, needs to incorporate multiple perspectives and seek to discover and negotiate points of agreement. The choice of the multicriteria decision analysis is justified by at least two considerations: the ability to handle multiple objectives and viewpoints simultaneously; and, the rigorous formal structure, which establishes a common language for the mediation of the decision-making process. The article contains a brief presentation of the Prospectar Program, details on the construction of the model and on its analytical possibilities for establishing the order of technology priorities. From the point of view of the organization of the innovation systems, the existence of a common methodological language provided by the multicriterial approach can facilitate communication between different actors and make the options of orientation and strategies clearer for different organizations.

KEYWORDS

technology foresight; selection of priorities; strategy; knowledge application; technological priorities; multicriteria analysis; innovation systems; ELECTRE III

INTRODUCTION

This article describes a methodological proposal elaborated within the ambit of the Study Group on the Organization of Research and Innovation (GEOPI/DPCT/Unicamp). (1) The challenge put forward by the Prospector Program coordination board, in 2001, consisted of generating analysis routines for the data obtained by Prospectar for the selection of technological priority topics and, furthermore, to promote the public and private organizations' strategic use of the information collected by the Program.

The main idea governing the work was to join methodological approaches from technological prospection (including forecasting and foresight techniques) and from the decision-making systems in order to create a useful foundation for the organization of innovation systems. There is therefore a fundamental conceptual motivation, preoccupied with the improvement of the innovation systems and the networks of actors who sustain them. This preoccupation is also present in previous works, which form the basis for the present methodological proposition. (2)

The choice of the multicriteria decision analysis is justified by at least two considerations: the ability to handle multiple objectives and viewpoints simultaneously and the rigorous formal structure, which establishes a common language for the mediation of the debate (or dispute) that accompanies the process of decision-making under these circumstances (based on what is recommended by Keeney and Raiffa, 1976 and Barba-Romero and Pomerol, 1997).

The selection of priorities, particularly when conditioned by prospective perspectives, is a necessarily complex and multifaceted problem. The expectations of the future are usually contingent and full of uncertainties. The different actors' notions of the future are influenced by their different contexts and perceptions. The sets of information and knowledge to which each individual has access are different. In Brazil, when it comes to subjects associated with science and technology, specialists' perceptions tend to be rather categorized by the very logic the development of these activities have presented in this country. Excessively dissimilar interpretations in the present will also lead to divergent perspectives in the future, so this must to be carefully dealt with in order to reach successful results from the prospective effort.

The challenge of defining priorities on the basis of a thorough prospective exercise, as is the case of Prospectar, should therefore incorporate these multiple perspectives and seek to discover and negotiate points of agreement. Therefore, right from the start, we reject simplistic approaches and those that seek to discover an objective and ideal solution. The nature of the problem is different. There are no objective priorities. The best decision is determined by the value systems that lead to it and the best will not be the same if the value systems are different. Although a total and categorical polarization between value systems is improbable, it is always possible to find points of agreement. From the analytical point of view, in the worst of the cases, the consensual priorities for heterogeneous group of actors could be a narrow set of options. In the best case, the priorities will be identified in those set of technological topics whose assessment is in concordance with a common value system that could reflect the future strategies of government politics, research institutions, enterprises or individuals.

The article continues with a general presentation of the Prospectar Program. The details on the construction of the decision model and its analytical possibilities for establishing the order of technology priorities are the subject of the proceeding chapter. The objectives of the article are to demonstrate the rationale for the proposed organizational approach, indicate future methodological research directions and discuss the appropriate use of the model.

PROSPECTAR PROGRAM

Beginning in 1999, the Prospectar Program, projected and implanted in Brazil, is the largest and most comprehensive prospective study performed to date in this country. The project involved the participation of almost 11.000 people and 16 'anchor' institutions, among these, research institutes and governmental regulation agencies.

The Prospectar is in consonance with the worldwide tendency of studies on technological prospection known as technology foresight (see, for example, Gavigan and Scapolo, 1999; Linstone and Grupp, 1999, and Johnston, 2002, for a introductory review on this subject) , which emphasis is on participation and the collective construction of the future. The use of Delphi as orienting technique of the prospection process provides the inclusion of a great number of participants and establishes a communication channel between them. This effect alone is already considered a positive aspect of the exercise. The strategy was to promote the circulation of information and simultaneously mobilize thousands of people involved in research and innovation to reflect on what would be relevant for the future. It was hoped that this would intensify the awareness of the participants about their role in this future, facing the many challenges established for the country, the propagation and circulation of new technological possibilities and a higher participation in the definition of the politics and decisions on the priorities in public support for science, technology and innovation.

The Delphi technique is employed as an interactive group discussion at distance and aims at the attainment of agreement through rounds of questionnaires with controlled reply delivery. A set of technological topics, selected in advance, is sent to a large number of potential respondents, and questions concerning these topics are made whose answers are also, in general, previously standardized in statistically treatable scales. After each round, the results are analyzed and the questionnaires returned to the respondents. At this point, each participant has the opportunity to review his or her questions in view of the consolidated opinion of all the other participants. His or her anterior evaluation can then be altered or maintained. This controlled reply process is brought to a conclusion when the level of stability obtained is considered satisfactory by the project coordinators or when the number of participants decreases too much (Rowe and Wright, 2001).

In the Prospectar Program, three Delphi rounds were carried out. The study focussed on eight main areas for which the anchor institutions prepared the lists of technology topics in order to initiate the Delphi. The topics should include the most important technologies for each area and make explicit their degree of maturity by beginning the description of each topic with the words 'Elucidation', 'Development', 'Use' or 'Wide Use'. The anchor institutions were also responsible for indicating participants to answer the questionnaire in their respective areas. The only criterion required for participating was a university degree. The participants were mainly selected on the basis of the so-called Curriculum Lattes Base (that collects the curricula of almost every Brazilian researcher working in the public sector or at universities) from The National Council for Scientific and Technological Development, known by its acronym in Portuguese--CNPq (http://lattes.cnpq.br). The lack of a database of the researchers and managers of innovation in the private domain rendered impossible a greater participation on behalf of these actors (who represented about 6% in the second round and 5% in the third).

The main areas considered in the study and the respective anchor institutions were: Farming and Cattle Raising (Embrapa), Aeronautics (CTA), Energy (Cenpes, Cepel, CNES), Space (AEB, INPE, CTA), Materials (INT, CETEM, INPE, CenPRA), Hydric Resources (CPRM, ANA), Health Care (Incor, Fiocruz), Telecommunications and Information Technology (CPqD, CenPRA, Socinfo) (see Table 1). It is important to note that the Environment and Biotechnology were included transversally among the areas.