General metrological inspection planning is among the least explored computer-aided process planning (CAPP) domains. This paper explores certain basic issues involved in inspection planning using case-based reasoning in an environment of a Generic CAPP Support System. Firstly, algorithmic methods for characterizing and extracting inspection features are proposed and discussed. A sequential knowledge based filtering method is developed to reduce the number of inspection features typically encountered in metrological inspection planning. Finally, a formalized approach for case representation of relevant inspection domain knowledge using a newly developed parametric-list technological feature graph (PLTFG) is presented.
Concurrent engineering (CE) has emerged as a commonly accepted solution to the sharply decreasing time-to-market problem in manufacturing. In practice, the effectiveness of CE depends greatly on the availability of highly automated and reliable computer-aided process planning (CAPP) tools. As a result, a variety of algorithmic and artificial intelligence (AI) based methods have been developed in the last two decades to address different problems arising in CAPP [1], [2], [3], [4], [5], [6], [7], [8] and [9].
The manufacture of any part/product involves several processes: machining, casting, injection moulding, inspection, etc. Under each of these processes, there are several sub-processes each with very distinct characteristics. For instance, machining may involve a variety of operations such as turning, milling, and drilling carried out on separate machines or on a single machining center. Likewise, inspection could be carried out through a sequence of operations each carried out with the aid of a different metrological instrument or on a single coordinate measuring machine (CMM). Further, one needs to evaluate several competing manufacturing processes at each stage in process planning. All this suggests that there is a strong need for a comprehensive CAPP system that encompasses most of the commonly found manufacturing processes in a seamless and integrated fashion. However, it appears that, in general, the currently available CAPP solutions do not meet this criterion. Firstly, the existing CAPP systems address very few of the commonly used manufacturing processes. While machining and assembly processes have been extensively supported in CAPP systems, many other manufacturing processes and inspection (with the exception of inspection using a CMM) have received very little attention. Secondly, since the planning of a manufacturing process requires reasoning based on process-specific technological knowledge, existing CAPPs tend to be fragmented rather than integrated with each process being dealt with by a totally independent module.
The issue of fragmentation of CAPP was brought into focus recently addressed by Yuen et al. [10]. It was suggested that, while each CAPP domain might be distinct in terms of the technological knowledge, the competing manufacturing processes at each stage of CAPP have one thing in common—they have to perform in only with almost the same part/product objective. It should therefore be possible to aggregate the reasoning processes related to diverse manufacturing processes into a common platform called the Generic CAPP Support System (GCAPPSS)—see Fig. 1—that precedes the individual process specific modules. Yuen et al. [10] suggested that the downstream process-specific modules could be integrated through a common technological reasoning strategy and that case-based reasoning (CBR) is a good candidate in this regard.
Full-size image (47 K)
Fig. 1.
The role of Generic Computer-aided Process Planning Support System in comprehensive computer-aided process planning.
Figure options
However, the implementation of the strategy outlined in Fig. 1 requires that we already have a fair understanding of the product-based reasoning strategies in the various CAPP domains. Unfortunately, at the present the degree of understanding is quite uneven across different CAPP domains. Amongst the most studied seems to be the domain of machining while the least studied seems to be that of inspection.
This paper aims to provide a deeper understanding of the product-based reasoning strategies required in implementing a computer-aided inspection process planning (CAIPP) system based on the notion of GCAPPSS. We will focus mainly on dimensional inspection of parts containing polyhedral and cylindrical features that need to be inspected by using a variety of commonly found metrological equipment. The next section will present an overview of the literature related to CAIPP. This will be followed by a brief review of dimensional inspection. The intention is to arrive at a reasonably complete and generalized set of observations regarding dimensional inspection and the logical characterization of inspection features.
Amongst the various CAPP domains, notwithstanding its enormous importance in industry, non-CMM-based inspection process planning has attracted very little research effort so far. This paper has tried to fill this gap by addressing two basic issues: inspection feature recognition and inspection case representation. This contribution underpins the future works directed towards inspection process planning utilizing CBR in CAPP environment. A series of domain-specific and knowledge-based filters have been proposed to contain the problem of inspection feature explosion.
A two-level hierarchical method for decomposing an inspection case into primitive elements has been described. By taking advantage of six or more orthographic projections of the part model, the approach enables information regarding setup and access to be included while describing cases. However, there would be occasions when the engineering drawing does not include all the necessary elevations so that some of the inspection features would not be visible and are presented in hidden lines. The hidden lines implies three possible situations: (1) the hidden features which are visible in the opposite direction, (2) the hidden features which are visible in the side direction, i.e. external/ internal undercut features, (3) the hidden feature which are not visible from all directions , i.e. closed cavity. In case of hidden features of cases 1 and 2, orientation and setting of the part or re-design of the inspection tool path in the inspection process plan to make the hidden feature visible could solve the problem. Yin et al. [39] applied visibility theory and developed algorithms for set-up workpiece and mould parting. Kweon et al. [40] utilizes the visibility map to solve part orientations for CMM inspection. For the case (3) of hidden feature, either redesign of part or application of non-touching measuring equipment such as X-ray, ultrasonic measurement could help. In further exploring these issues, it should be remembered that the reference frame should, in general, be the same as that used by the designer. Often the designer would take the functional direction of the part as the reference direction. Other times, he might adopt the manufacturing direction. Whatever is the designer's intention should be evident if one has an engineering drawing containing orthographic elevations or a solid model with a specified global coordinate system. Further work is needed to arrive at a robust solution. This paper provides a framework of the methods and algorithms in the “Represent” sub-system of an automated CBR system for Inspection Process Planning.
