Luis G. Occeña

A knowledge sorting process for a product design expert system

Knowledge sorting is one way to organize the knowledge acquired from the domain expert(s) and various sources. Well-organized knowledge representation will make rule extraction much easier. This paper describes a knowledge sorting process that was developed to facilitate the rule extraction for a product design expert system. The process capitalizes on the relationships between design attributes and factors, dependent and independent variables, and consists of three stages: identification of knowledge sources, generation of taxonomic trees, and organization of acquired knowledge. An example applies the sorting process to the development of an expert system for the design of wood head golf clubs.

CONDENSE: A concurrent design evaluation system for product design

In dealing with the issue of integrating design and manufacturing, concurrent design evaluation plays an important role. This paper presents a concurrent design evaluation system (CONDENSE) developed to help product designers in evaluating possible design solutions and design alternatives during the early stage of design. It consists of two important functions: a qualitative aspect evaluation applied during the stage of searching for combinations of solution principles to help determine the design specifications, and a quantitative aspect evaluation applied to provide information on performance, assemblability, manufacturability and costs to facilitate design selection. The framework of CONDENSE is based on a blackboard architecture that requires the classification of knowledge into appropriate knowledge sources. As design data are interrelated and may have uncertainty, a graph decomposition algorithm is used in constructing the knowledge sources, and a linguistic evaluation module is integrated with the qualitative aspect evaluation subsystem to deal with data uncertainty. The proposed system has been validated with respect to the design of golf club heads. The results, which have been validated by experienced designers, are promising and can contribute to the speed-up of design and development, improvement of design quality and facilitation of design selection.

Computer Integrated Manufacturing Issues Related to the Hardwood Log Sawmill

This paper evaluates the issues associated with the computer integration of the various technologies now available to hardwood log sawmills, so that these technologies will cooperate in helping the sawmill achieve better lumber yield. The concept of computer integration, and the current sawmill setup, are reviewed. A model of a computer-integrated hardwood sawmill is described. Issues concerning the integration of both hardware and software are also discussed.

Lumber value differences from reduced CT spatial resolution and simulated log sawing

Abstract In the past few years, computed tomography (CT) scanning technology has been applied to the detection of internal defects in hardwood logs for the purpose of obtaining a priori information that can be used to arrive at better log sawing decisions. Because sawyers currently cannot even see the inside of a log until the log faces are revealed by sawing, there is little perceived need to obtain scan images as detailed as those obtained in medical CT imaging. Yet, CT scanner speed and the usefulness of CT data for decision-making are dependent on the spatial resolution of scans. Spatial resolution is a function of three factors: physical pixel size, scan thickness, and scan frequency (pitch). A 3×2 3 factorial experiment was designed with two levels for each of these three factors, to test their effect on lumber values. Three hypothetical logs corresponding to three hardwood log grades were simulation-scanned, then simulation-sawed by a human operator using a modified Malcolm opening face heuristic. Log grade affected lumber value recovery as expected, although reduced spatial resolution (by doubling the pitch, thickness, and pixel size) exhibited no discernible pattern in our statistical tests for effects. Volume recovery for below grade boards was predicted very accurately by size, thickness, and pitch-size. The greatest opportunity for lumber value recovery improvement using information-augmented sawing appears to be in grade #2 logs.

Knowledge decomposition for a product design blackboard expert system

Abstract Product design is a complex process, requiring many design factors and knowledge areas to be considered simultaneously. An expert system with a blackboard architecture is appropriate for representing this variety of factors and areas. However, oftentimes it is unclear how to transform the knowledge into the component knowledge bases of a blackboard expert system. The paper describes a non-directed graph decomposition procedure for transforming the knowledge into knowledge bases as part of an approach for developing a product design blackboard expert system. A wood head golf club design example is provided.

NONDESTRUCTIVE EVALUATION OF HARDWOOD LOGS: CT SCANNING, MACHINE VISION AND DATA UTILIZATION

Sawing of hardwood logs still relies on relatively simple technologies that, in spite of their lack of sophisticatio n, have been successful for many years due to wood’s traditional low cost and ready availability. These characteristics of the hardwood resource have changed dramatically over the past 20 years, however, forcing wood processors to become more efficient in their operations. In spite of some recent advances, the breakdown of hardwood logs into lumber continues to be hampered by the inability of sawyers to “see” inside of the log prior to making irreversible cutting decisions. The need for noninvasive assessment of hardwood logs prior to breakdown is well accepted, but is difficult to realize because industrial scanning. in this context, is unique in several respects. For example, large volumes of material must be inspected quickly over an extended duty cycle, the wood material still possesses relatively low value compared to other industrial materials that require internal scanning, and many wood processors are small operations located in rural areas. Successful implementation of new scanning technology, however, will have tremendous payback for wood processors. and for timber resource conservation efforts. The research program reviewed here applies a three-pronged approach to address this situation. First, a relatively new and innovative CT scanning technology is being developed that can scan hardwood logs at industrial speeds. Second. machine vision software has been created that can interpret scanned images rapidly and with high accuracy. Third, we have developed 3-D rendering and analysis techniques that will enable sawyers to apply image assessment to actual log

A logic-based framework for address interpretation and rectification

Abstract To speed up the process of sorting letter mail, the U.S. Postal Service uses optical character recognition (OCR) machines on machine-printed mail. Even machine-printed mail can have missing words or characters in the address block or contain erroneous words or characters, resulting in rejection by the OCR as unreadable mail. The current procedure for dealing with such cases of patron errors is to subject the rejected mail to manual sorting. Missing character zip codes alone account for 61 percent of address line sets classified as unrecognized addresses. Manual sorting is tedious if left to unaided human perception alone. This paper describes a logic-based framework for address interpretation and rectification of machine-printed address mail rejected by the OCR for various reasons. A logic-based format was used consistently through the model, and implemented in Prolog. Preliminary testing of this early version of the model on 50 randomly selected samples of machine-printed mail rejected by the OCR at the Columbia general mail facility showed that 27 (54%) out of the 50 samples were rectified successfully. A classification of the types of errors in the study sample is also presented.

Pattern directed extraction and characterization of defect configurations in solid log models

Abstract This paper describes a knowledge based procedure for the automated extraction and characterization of internal defect configurations in hardwood log processing. Before sawmill automation can be realized, the capability to automatically extract and characterize internal defect configurations in each log must be provided. The procedure presented uses a five-stage filtering approach to extract the defect hull, and a three-stage characterization approach to describe the defect configuration type. The logic programming paradigm used to implement the procedure is described, and examples are given. The issues of computer extraction and characterization of internal log defects will find parallel relevance in the interface and integration efforts in computer-aided design and computer-aided manufacturing (CAD/CAM).

Knowledge organisation of product design blackboard systems via graph decomposition

Knowledge organisation plays an important role in building a knowledge-based product design blackboard system. Well-organised knowledge sources will facilitate the effectiveness and efficiency of communication and data exchange in a blackboard system. In a previous investigation, an approach for constructing blackboard systems for product design using a non-directed graph decomposition algorithm was proposed. In this paper, the relationship between graph decomposition and the resultant blackboard system is further studied. A case study of a number of hypothetical blackboard systems that comprise different knowledge organisations is provided.

GSS—A CAD-based graphic sawing simulator for hardwood logs☆

Abstract With the objective of studying the breakdown process for hardwood logs, a graphic sawing simulator (GSS) was developed to provide the capability for repeated sawing. Based on polyhedral solid modelling concepts and device-independent graphics, the GSS represents a solid log as a nonregular polyhedron and enables decomposition of the log (“sawing”) via regularized CSG Boolean operations.