CAPP - Three Approaches: Variant, Generative, and the Hybrid

Computer Aided Process Planning, (CAPP), is a production orginazation activity that determines how a product is to be manufactured. This is a cornerstone in the manufacturing process. A major part in determining the cost of components and the best combination of manufacturing tools and processes. This directly affects production efficiency, product quality, and company competitiveness. CAPP is a crucial link between design and manufacturing.

Even with today's technology and the everknown importance of process planning, this activity is still very labor intensive. One leans heavily on experinece and intuition that gives the required insight to the different manufacturing processes and possibilities within a company. However, the dependency upon intuition often eliminates a thorough analysis and optimization a process plan. This can result in delays and increased costs. CAPP takes on the role of standardizing processes, reducing generation time, and attempts to ensure consistant quality.

The Variant Approach
The traditional approach to process planning. Here, a part drawing is examined and then similar parts produced in the past are identified. For the previous parts, process plans are examined and then modified to suit the new part at hand. The primise: Similarities between certain producs imply that those parts can be manufactured in more or less the same way. This is the exact route that manual process planning, involving intuition, takes. Computer logrithm strenghtens the identification of 'families' with codes and definitions. Within small corporations, such 'families' of parts are not difficult to identify. The disadvantage of variant planning is that the process plan only caters for a rough outline plan that still needs to be adapted. The main drawback with this approach is that constructing computerized classification systems have yet to be found error free.

Generative Planning
With the generative planning approach, the computer program incorporates metal cutting know-how and the geometric vision of the part. The process plans are generated by making use of algorithms, decision logic, formulas and geometry based data to perform unique processing decisions to take the part from raw material to a finished state. Here there is no referral to previous plans. Part specificatons, as you can guess, are mandatory input. This includes variables such as material for example. After that, the process is fully automatic and it produces plans of consistant quality. However, the systems are complex and difficult to develop. Maintenance is very difficult for there are a large database of rules which have to be consistant in all condidtions. This results in higher probabilities of system failure.

Hybrid Planning
Introduced to limit the drawbacks of both planning approaches, and benefit from there advantages. Generative processes produce consistant results without classification on high levels of complexity. Variant processes are simple and easy to maintain.
The first step in a hybrid program: the workpeice is associated to a family. Associated with these families is a knowledge base that contains all possibilities to manufacture the part. The user again defines mandatory variables; the program attempts to generate an efficient, cost saving plan for the involved processes.