Figure 1. DFM process: Identify, validate, conversion
The DFM Process outlines the steps necessary to prepare DFM price data for discussions regarding purchased part prices with suppliers. The starting point is compiling a list of parts with the highest spend in the supply chain. This is simply gathering a list of parts with the standard price and yearly volumes for each to determine the highest amounts of spend. Typically, Dynisco uses part pricing and volume, as well as the typical batch or lot sizes combined with the 3D model and part drawing to run the DFM analysis. The batch sizes are usually what quantity the part is purchased in by the supply chain each time production requires the part.
Then, take a manufactured part from the supply chain, along with its purchased price, and run a DFM analysis of the part to determine if the supplier’s pricing is reasonable. A side-by-side comparison of the standard price and the DFM price either will validate the purchase price or show a gap in pricing. This is the “Identify” phase in the DFM Process. Parts with differences between standard and DFM pricing signal a potential for a cost savings.
The “Validate” phase can take multiple meanings. First, DFM pricing that is approximately equal to or greater than the supplier’s price will validate that the supplier is charging a suitable price for the part. Second, if a large price gap exists between the supplier and DFM price, it is incumbent upon the DFM user to validate the accuracy of the cost model, as well as investigate the print specification of the part to determine design features and tolerances that can be defined as the cost drivers.
The third and final phase of this DFM Process, “Conversion,” is to use the data generated from DFM analyses and engage suppliers to negotiate lower prices for purchased parts—essentially converting differences in price into savings.
Figure 2. Material cost in DFM
For an example, we use a part machined on a lathe. Once the approximate envelope dimensions are entered into DFM and the process and materials are selected, DFM will first assess the material costs for the part. Next, the batch (lot) size is entered into the program (Figure 2). This will have the most impact on the setup cost for each machine used to process the part. It will also impact any special costs amortized across the volume of parts such as tooling, fixtures, or programming.
The software provides these costs as defaults for more common materials, such as steel or aluminum. The exact cost per pound can be adjusted if known from the supply base or when using exotic or expensive materials such as Inconel. Rounding out the critical inputs for estimated material costs, are the volume and weight of the final part geometry and the overall dimensions for the work piece of the part. The workpiece is defined as the envelope dimensions of the stock geometry of the material being machined to ensure there is “extra” material to start with before machining begins.
Figure 3. Setup cost in DFM
Once the stock process for the material is defined, the user specifies what machines are used to process the part. The DFM software contains an extensive database of various machines. For this example part, a generic CNC turret lathe is used. Under the setup/load/unload row for each machine used in an analysis (Figure 3), the number of part reversals (how many times a part is moved within the setup), setup time, and again, tooling, fixtures, or programming can be entered in this row.
Figure 4. Processing cost in DFM
Then, machining processes are entered from the different operations within DFM’s Operations Library (Figure 4). The machining operations can be defined by entering dimensions for material removal. This data can be easily calculated by using the part print and starting work piece geometry. Reject rates are entered as a percentage for each machine selected for a DFM analysis.
Figure 5. “Extra” operations in DFM
Extra operations such as part inspection, measurements, and cleaning should be used consistently throughout the DFM process. While these smaller operations add minimal costs to the part price, it’s good practice to show suppliers the level of detail used in creating DFM analyses.
Additionally, and more important, if a part has finish requirements such as heat treatment, plating, or painting, those can be included too. Shipping requirements such as bagging and boxing of parts should be considered also. Lastly, supplier overhead and profit should be taken into account to complete a full DFM price analysis (Figure 5).
Once the DFM analysis is complete for a list of parts from the supply chain, there are multiple ways to prepare pricing results for discussions with suppliers. The DFM software offers reports that can be exported from the software, such as an Executive Summary, a Results Summary, and a Totals Summary. Dynisco prefers to use the Results Summary, which gives a line-by-line display of every operation in the DFM analysis, while breaking down the processes time and cost for each. This summary allows users to match the operation or machining process to the equivalent dimension on the print.
Figure 6. Chart equating surface roughness to tolerances.
Part prints should be reviewed during the DFM process to understand how tolerances drive processing time and impact manufacturing costs. Many machining operations in DFM ask the user to specify the surface roughness of the part. The DFM software equates surface roughness to approximate tolerances, which in turn, directly relates to the amount of processing time required. Processing time translates into the manufacturing cost. The message is to be aware of how tolerances can impact cost, as shown in Figure 6.
Figure 7. Total spend on manufactured parts
This example uses the total spend on approximately 350 parts across four Roper business units from 2013. To start, the parts list with standard purchase price and yearly volumes have been prepared. The total spend on DFM analyzed parts from each business unit will look like the chart in Figure 7.
Figure 8. Comparison of total standard cost spend
Next, the DFM generated prices are multiplied by the same part volumes and compared to the standard cost total spend. This shows differences between standard and DFM prices and where the opportunities are for potential cost savings (Figure 8). The DFM prices were higher than standard at business units #2 and #3. In this case, the DFM analyses are “validating” suitable pricing for parts at these two sites.
Figure 9. Potential savings at business units #1 and #4
Another way to look at the areas for potential savings is shown in Figure 9. The yellow bars show the opportunities for potential cost savings against the total spend on the parts. In this example, DFM prices were collectively 52 percent lower at business unit #1 and 10 percent lower at business unit #4. These are the parts that require discussions with the suppliers who provide them.
Figure 10. Realized savings at business unit #4
Figure 10 shows the cost savings results for 2013 at the four business units being examined. The bar in green represents the value of improved purchased price variance at business unit #4.
Figure 11. Monthly and total savings at business unit #4
The scale of the chart in Figure 10 does not demonstrate the significant results that were accomplished. In Figure 11, displaying month-to-month results, illustrates that a realized savings was accomplished each month of 2013. Additionally, that steady stream of cost savings continues into the first four months of 2014. These results are directly related to conducting meetings with suppliers as previously described. The partnership between business unit #4 and a key supplier was solidified. Because of the willingness to participate in the DFM program, this key supplier is now being presented opportunities to quote parts from other business units.