We describe a design problem arising in the color printing industry and discuss a number of integer linear programming and constraint programming approaches to its solution. Despite the apparent simplicity of the problem it presents a challenge for both approaches. We present results for three typical cases and show that the constraint programming approach provides better results, although in some cases after considerable handcrafting. We also show that the results obtained by constraint programming can be improved by a simple goal programming model.

This paper summarizes a biological key using a fuzzy logic expert system. This approach allows the user to make some errors, be uncertain about some answers and not answer questions if unsure, but have an opportunity to identify unknown specimens correctly. This approach was adopted in order to develop a ‘forgiving’ key for users with little or no prior knowledge of identification. The system was implemented using a multi-media expert system builder, ‘Matcher’, operating in a Microsoft Windows environment. The paper reports the design issues for a key to identify freshwater invertebrates to family level and describes the types of questions used within the key, the fuzzy inference engine, the structure of the system and the methodology adopted for testing it. Following this, an empirical test is reported of the hypothesis that a fuzzy logic expert system can improve identification accuracy for a sample of intended users. Finally, the discussion considers the practicalities of using such a system as an aid to organism identification. Information regarding the software is available on request from the authors. Inquiries relating to the key should be sent to Linton Winder ([email protected]), whilst inquiries regarding the expert shell ‘Matcher’ should be directed to Martin Lefley ([email protected]).