معاملات فاکتورگیری حمایتی در برزیل با استفاده از نقشه های استدلال: DSS مبتنی بر زبان برای ارزیابی حساب های دریافتنی

Factoring companies are a widespread way of providing working capital to small enterprises in Brazil. This type of financial transaction has higher risks when performed in developing countries, due to unreliable financial information on firms, an unstable environment, and particular managerial practices. This paper describes a case study in which a language-based DSS was developed for a Brazilian factoring company to evaluate the perceived risk of buying accounts receivable; and discusses the suitability of different approaches to decision support for this type of decision in Brazil—which may be relevant for similar situations in other developing countries.

Factoring companies are a widespread way of providing working capital for small enterprises in Brazil. A factoring transaction is a triangular financial operation, where the factoring company buys accounts receivable from its client (the creditor firm), and takes the burden of collecting them from its client's customers (the debtor firms). Brazilian factoring companies operate a high risk/high return business, in an environment characterised by high interest rates and spreads, an unstable macroeconomic situation, a weak legal framework which does not allow a quick recovery of collaterals, and a lack of adequate information on potential borrowers (see Ref. [9] for details on this problem). Perhaps surprisingly, given the prevalence of these operations and the general interest in credit appraisal techniques, there appears to be a lack of decision support systems developed for the tri-partite factoring transactions discussed here. This paper is part of an ongoing action-research effort which attempts to support decision-makers dealing with this type of problem (see Ref. [4] for a discussion about the use of action-research in the DSS context). This has the dual aim of providing decision support and exploring the appropriateness of different approaches to doing so, given the distinctive characteristics of the problem context. In a previous case study, we employed a quantitative multi-criteria model, which evaluates several dimensions of perceived risk, for developing a DSS (see Ref. [5] for details). This first intervention provided several insights into particular characteristics of this problem, especially its qualitative nature and the decision-makers' willingness to think and talk in qualitative terms. These insights lead us to intervene again; now using a different, recently developed decision tool—called a reasoning map—which provides integrated support for both problem structuring and qualitative assessment of alternatives using a cognitive/causal map structure. Reflecting the nature of the action research, the paper is aimed at DSS practitioners and researchers with an interest in the specific application area of financial decision-making and/or the appropriateness of different approaches to decision support for this type of decision in developing countries. The paper begins with a brief introduction to reasoning maps, which is then elaborated through its use in the development of a DSS to evaluate accounts receivable. We conclude with reflections on the process, in particular on how the specific characteristics of the problem may influence the use and development of DSSs.

This paper has described the development and application of a DSS based on a recently developed and evolving approach to multicriteria decision aid, referred to as a reasoning map, which provides an integrated framework for problem structuring and evaluation. The problem considered, the evaluation of accounts receivable, is one that is faced on a regular basis by Brazilian factoring companies. The aim of the intervention was neither to provide a normative rule of choice nor a descriptive model which reflects unaided decision-making. The objective was to enhance the decision-maker's understanding of the issue, by taking him through a structured process which encourages hard thinking, and by providing a tool that may make him more aware of the evaluation of a given account receivable in terms of his ultimate goals, through multiple and interacting chains of influence. We also have drawn reflections from two interventions on this problem, regarding some challenges and opportunities in designing DSSs for this type of problem in developing countries. We believe that more research involving the development of DSSs in real-world interventions in developing countries is needed. Future research may address different types of decisions in the same country using the same decision method (therefore extending in-country generalizability of outcomes), or the same type of problem in different countries using the same decision method (thus enhancing cross-country generalizability of conclusions). Another interesting future research would be using different decision methods for this factoring problem like, for example, Expert Systems, Fuzzy Cognitive Maps or Bayesian Networks. An important aspect is, in our view, that the research design of these interventions focuses explicitly on how the developing countries' specific features may influence the choice of a decision method and on the DSS design.