|
...manufacturer's defect rate is sufficiently low without it. Unnecessary improvement may in turn cause significant delay to production for the manufacturer and ultimately increase the price of supplied parts. Such is often the case for automobile body manufacturers during die tryout, prior to vehicle launch. This paper presents a formulation for quantifying the time saving potential of an approach to die tryout known as "Functional Build" (FB).
In the automobile sheet metal stamping industry, die suppliers test and rework newly constructed dies during a pre-production "die tryout" period. New dies commonly produce panels with dimensions biased away from target because of uncertainty during die design. Centering panel dimensions on target often requires extensive die rework. However, evidence suggests that much of the die rework is unnecessary for producing panels that result in dimensionally correct assemblies at the next manufacturing stage (Hammett et al., 1995; Hammett et al., 1999).
The problem of unnecessary die rework, comes from a lack of two kinds of information during product design. First, panel designers may be uncertain about the difficulty of building dies to meet panel design. Consequently, the specifications on two mating parts may result in extensive die rework for one supplier and none for another. Second, designers may lack understanding of the dimensional relationship between the panels and the assembly. Attempting to guarantee no assembly defects, they often assign unnecessarily narrow specification limits.
The FB solution is to build assemblies from the out-of-spec panels and identify the least costly die rework needed to achieve a functionally acceptable assembly. In contrast, the traditional purchasing agreement requires the die supplier to refine its processes until meeting specifications with little or no communication with the OEM. Consequently, opportunities to alter part designs and avoid unnecessary improvement are missed.
Recent studies of FB in the North American automobile industry suggest that FR has reduced die rework costs, die tryout times, and improved product quality for some firms (Hammett et al., 1995; Hammett et al., 1999). Yet, FR requires a lot of time and resources and the benefit of FB depends on many factors including panel and subassembly characteristics and die supplier expertise. The model presented in this paper yields a comparison between die tryout with and without FR in terms of the expected die rework duration.
2. Functional build and net build
An automobile body is comprised of 200-300 sheet metal panels, each stamped from a single sheet metal blank. The blank goes through a sequence of four or five dies, called the die set, that perform draw, trim, punch, and bend operations. Because of variation in the stamping process, panel dimensions are random variables. The mean deviation from target of a panel dimension is called the "bias". Removing dimensional bias usually requires reworking the die set to change the geometry of all or some of the dies.
Dimensional bias is prevalent among all world automobile manufacturers (Hammett et al., 1998), and almost all die sets require rework before use in production. The prevalence of dimensional bias indicates the uncertainty inherent in die design. Die engineers must predict the panel dimensions that will result from a given die geometry; a skill (often called an art) that requires experience with metal flow and spring brack in the stamping operations.
Die rework entails removing the die set from the press, dismantling the dies, welding to add metal, and machining for the desired shape. After machining, the dies are re-assembled, stoned for a highly smooth finish, and mounted on the presses for more testing. Die rework times range from 1 week to several months depending on die complexity and the number of dies in the set requiring rework. Consequently, die tryout adds significant time to die development, which constitutes a significant portion of the total time spent prior to the launch of a new vehicle (Clark and Fujimoto, 1991).
Even though die design uncertainty and dimensional bias are common among all automobile manufacturers, some spend far less time in die tryout than others. Several studies show that North American manufacturers required significantly more die tryout time than Japanese manufacturers (Womack et al., 1990; Clark and Fujimoto, 1990). Short Japanese die tryout times have been attributed, in part, to the practice now known as FB. North American manufacturers later adopted FR in the early 1990s (Ward et al, 1995).
Each manufacturer takes a different approach to FB, but the basic method is the same. After die construction, the die supplier submits sample panels and panel dimensional data to the manufacturer for building and evaluating a "screw body" assembly, a manually constructed assembly for testing panel fit. The name "screw body" comes from the common use of screws in joining the panels. (See Hammett et al. (1998) for a complete description and history of FR.) The screw body assembly is then examined by a group of engineers from product design, die design, stamping, and assembly. Based on expertise and die rework cost and time estimates, the group of engineers decides which assembly dimensions to correct by reworking dies. For those dimensions with a large bias that require no die rework, the specification limits may be widened and the design target may be shifted.
In contrast to FB, the traditional "Net Build" (NB) approach to die tryout requires the die supplier to rework dies until the fraction of defective panels (those having out-of-specification dimensions) is below a given threshold. The panel specifications and the threshold for defect rate are fixed before die tryout. Thus the NB die rework policy is determined before the biases are known, and the panels in a subassembly are treated independently. In FR, the die rework policy is determined after the biases are known and the panels are treated jointly.
Before introducing the model we briefly examine the extant literature on analytical models for coordination in product and process design. Krishnan et al. (1997) model the exchange of preliminary design information from an upstream activity (product design) to a sequentially dependent downstream activity (product or process design). Loch and Terwiesch (1998) model the overlap and the communication policy between two sequentially dependent development tasks. Both formulations relate to die tryout in the sense that panel design changes are sometimes transferred to die design during die tryout. In contrast, a FR die tryout entails panel changes originating in die tryout and being transferred to panel design; i.e., there is an interdependent relationship between the upstream task (panel design) and the downstream tasks (die design and die tryout).
Ha and Porteus (1995) model the exchange of design information between two interdependent design activities in review meetings where design flaws are identified. Their formulation relates to a FB die tryout because the FR meeting is a review between product and process design engineers. However, die tryout is not a design stage. Design flaws are not created in die tryout, they are only identified. The review meeting is one-sided in the sense that the downstream activity is reviewing the designs (panel and die designs) of the upstream activities.
3. The model
To...
NOTE: All illustrations and photos
have been removed from this article.
More articles from IIE Transactions
Brokering the customer-supplier partnership in product design and real..., April 01, 2003
Looking for additional articles?
Search our database of over 3 million articles.
Looking for more in-depth information on this industry?
Search our complete database of Industry & Market reports by text, subject, publication
name or publication date.
About Goliath
Whether you're looking for sales prospects, competitive information, company
analysis or best practices in managing your organization,
Goliath can help you meet your business needs.
Our extensive business information databases empower business
professionals with both the breadth and depth of credible,
authoritative information they need to support their business
goals. Whether it be strategic planning, sales prospecting,
company research or defining management best practices -
Goliath is your leading source for accurate information.
|
