Insurance fraud evaluation. A fuzzy expert system

All the studies dealing with the Italian Insurance market, show that fraud is a raising, relevant problem of that sector. ISVAP results (Istituto per la Vigilanza Sulle Assicurazioni Private e di Interesse Collettivo) stressed the remarkable number of around 200.000 fraudulent claims in Italy for year 1998, 155.550 referring to the lone car sector. This occurrence leads the insurance market to higher costs for honest customers and to a low efficiency of the whole system. To solve the problem, ISVAP itself is going to create a Database of actors, vehicles and fraud patterns to provide Companies with a tool to easier investigate on suspicious cases. On their side, Companies are trying to embed real “Fraud Units” in their typical activity, to identify suspicious cases and fraudulent patterns either in the insuring phase or in the settlement of claims. All the efforts to face the problem aim to redraw the issue as an industrial cost problem to handle general costs, distribution costs and claim compensation with efficiency principles. Against a deeper, more efficient investigation activity, Company shall face three opposite problems: high cost of expert activity, the request of fast settlements and, for Italian market, the requirement to cover any people who ask for a Policy. Fraud investigation and fast settlement of claims are antithetical activity. Call centres well reply to speed request, but let unsolved or even worsen the analysis on suspicious claims. The low competence of call centres employees doesn’t allow to leave them a first level of judgement on the genuine nature of claims. Thus proceeding, any audit analysis requires fraud experts. That’s the reason why fast settlement tends to generate extra costs in fraud investigations. What Companies need is a standard, automatic, fast control method to filter real suspicious cases to fraud experts and let Call centres free to pay immediately the majority of claims. Unsuspicious claims can thus be settled automatically, even by the non-expert call centre operators, while claims that exceed a fixed threshold value will be investigate by fraud experts. Claim auditors can so dedicate their activity to potentially fraudulent claims only. The aim of this paper is showing how a Fuzzy Logic Control (FLC) model can efficiently evaluate an “Index of suspect” on each claim, in order to stress fraudulent situation to be investigated from the experts