Probabilistic expert systems for managing information to improve services

Purpose – The purpose of this paper is to propose the use of probabilistic expert systems (PES), as a tool for managing complex multivariate and highly structured information. The aim of this paper is to verify PES' effective potentiality in quality management. Design/methodology/approach – In the social surveys context, the questionnaire is a well‐known tool for gathering data. Statistical techniques characterised by different levels of complexity can be used to extract from the collected data differently detailed information to support decisions. The paper focuses on PES, belonging to the family of multivariate statistical models, that are presented by an application to a public administration survey. Findings – The results of an application of PES to a citizens' satisfaction survey are presented. These results will show that probabilistic expert systems are a promising tool for service improvement analysis, based on customer perceptions. The key factors that have an impact on overall satisfaction, suggesting potential improvement areas in processes are identified in detail. Practical implications – PES can integrate subject‐matter knowledge and statistical information obtained from the questionnaire producing a knowledge instrument. Having this kind of knowledge helps one to make managerial decisions and plan improvement actions. Originality/value – PES can be considered as an innovative and valid way to orient strategic decisions. In particular, using the information enclosed in PES and the know‐how concerning the organization, the decision‐maker can take decisions supported by a scientific and objective tool.