7.3.1 Product Structure Graph
The product structure graph (PSG) illustrates the product variety in a hierarchical tree and allows engineers to focus on critical features. This will allow them to identify and eliminate unnecessary varieties. It also assists in determining which components are considered as standardized and which subassemblies are to be modularized. The inputs to the PSG should be the scope of components and features within the product line. For example, the inputs could be from QFD analysis (Martin and Ishii 1996) or include product permutation and the combination of these permutations that creates the different product variants. The output of PSG
75–85%
15–25%
Variety Cost 0 %
100% Total Cost
20–40%
5–10%
30–40%
10–20%
10–20%
Product Development
Purchasing
Production
Logistics and Transportations
Operations
Figure 7.2 Impact of variety on total cost.
is the product structure, including variants. Designers could use this information to better design and achieve optimum design combinations. The major goal of using the product structure graph is to optimize the manufacturing process used for producing the variety in the given product line, while minimizing the invest- ment required. The graph provides only a qualitative guide to the design of varieties.
The product structure graph and complexity measures were applied to a variety of problems, including automotive window regulator, heat trace cable connectors, and hard disk drives (Martin and Ishii 1996). The graph and the complexity measures will clarify the overall product structure, identify cost drivers, and provide a visual guide to redesign opportunity.
7.3.2 Process Sequence Graph
The process sequence graph illustrates the flow of the process sequence and its differentiation points. Differentiating the product later in the assembly process will reduce inventory costs and the complexity of the manufacturing system (Lee and Billington 1993). These strategies depend on the manufacturing time with respect to the required lead time (Martin and Ishii 1997).
7.3.3 Commonality Graph Method
Another method used to manage the variety and its cost is the application of the commonality graph method. In this method, a series of charts are developed
Indirect Impact Direct Impact Part Variety
Product Variety
Process Variety
Operations Set Up
Inventory Management
Material Handling
Stability of MPS
Supporting Services
On-time Delivery
Finished Quality
Total Costs
Product Flexibility
Figure 7.3 Impact of product variety on the manufacturing system. (From Yeb, K., and C. Chu. 1991. International Journal of Operations and Production Management 11(8): 35–47.)
based on the selected industry that correlates the commonality of components to (Martin and Ishii 1997)
◾ Process sequence
◾ Lead time of components
◾ Amount of variety desired by customer
7.3.3.1 Process Sequence versus Commonality
The relationships between the commonality of features and the process sequence are presented using this graph. A commonality index for each component (CIcomp) is calculated as
= − −
CI 1 −1
comp U 1
Vn (7.1)
where
U = number of unique part numbers Vn = final number of varieties offered
If only one component is considered sufficient for all the required varieties, the CIcomp is set to 1. This is considered the desired options.
7.3.3.2 Lead Time versus Commonality
Standardization can be eliminated if there is a low level of commonality and short lead time among the components. If there is a long lead time, 100% commonality is desired in order to minimize the inventory costs associated with the safety stock level and to establish standardized components.
7.3.3.3 Customer Requirements versus Commonality
Variety voice of the customer (V2OC) is used as a measure of importance for identifying the component variety demanded by customers. It represents the impor- tance of the component to the customers as well as the heterogeneity of the market.
One of the techniques to measure that attribute is through conjoint analysis.
7.3.3.3.1 High Variety Low Volume (HVLV)
The principle of lean manufacturing (LM) is modified to match the high variety and low volume (HVLV) conditions (Jina et al. 1997). This framework is illustrated as shown in Figure 7.4.
7.3.3.4 Design for Logistics and Manufacture (DFLM)
DFLM is critical for HVLV since it will reduce the cost and complexity associated with adding variety through
◾ Common raw material parts
◾ Common finished parts
◾ Modular designs
◾ Staged engineering change control
◾ Multifunctional teamwork
7.3.3.5 Organizing for Lean Manufacturing
Organizing for lean will reduce the variation in the material flow. This could be achieved through organizing the high-level demand (assembly and subassembly) and integrating the customer demand stage and the order release stage.
7.3.3.6 Integrative Supplier Relationships
This method proposes the use of generic raw material design, part, and subassem- blies from single source rather than multiple vendors.
7.3.3.7 Process Orientation and Consistent Performance Measures
This measure is used to monitor the operation’s progress. For example, the measures of HVLV situation includes batch sizes, space utilization, setup times, numbers and the justifications for unplanned engineering changes, supplier delivery frequency, customer satisfaction ratings, and delivery time.
Design for Logistics &
Manufacture
Organizing for Lean Manufacture
Developing the Supply
Chain Agile Processes
& Consistent Measures
Figure 7.4 Adapting LM principles to HVLV.
7.3.3.7.1 Empirical Implementation
Empirical approach has also been used to measure the impact and required varieties.
One was based on five case studies done in Britain and Brazil by Da Silveira (1998).
The primary goal was to develop a product variety management framework. Results from these five case studies (practical results) and literature reviews ( theoretical) suggested the framework as illustrated in Figure 7.5.