In our previous blog post, we described three approaches to answer the “how much will it cost” question without having to RFQ the supply base or adopt complex clean sheet estimating software. These simple methods allow you to estimate costs based on physical part attributes and volume data. In this post, we outline how you can calculate even more accurate cost estimates for those commodities where you are getting supplier cost details.
Join APD for our upcoming webinar on March 22nd.
Click here for more details.
Why should you go to the trouble of developing more accurate should-be cost models? In some cases, for example when you are estimating the cost of a simple design change that primarily affects material weight, you probably shouldn’t. But if you plan to use cost estimates to negotiate pricing with your suppliers, a more thorough understanding of their processes and costs will give you the credibility to have fact-based discussions that we find lead to better outcomes.
Manufacturing process cost models are developed using cost drivers: those factors that drive variable costs. The simplest cost drivers – material, labor, and overhead – can be determined by analyzing data from suppliers’ detailed cost breakdowns from past quotes. A few tips for getting usable data from suppliers:
- Configure your detailed cost breakdowns to support cost models – Typically this means using commodity-specific formats and including physical part attributes
- Be careful to differentiate costs by region – Costs vary considerably by country and often times by region in the country (ex: manufacturing labor costs in the Southeast vs. Midwest United States)
- Promote suppliers providing true cost information with their quotes – communicate your expectations and question inconsistencies
- We typically choose the 25th percentile value for most cost drivers (minimum is not realistic, and average is not aggressive enough)
Machine size, cycle times, and material usage are examples of cost drivers that can be obtained by analyzing physical attributes (cost tables, linear price models, and multivariate regression of data from past supplier quotes). For more information on this, click here to download APD’s Step-By Step Guide: Using Physical Part Attributes to Create Should-Be Cost Models.
Below is an example of what the process labor and overhead burden costs look like from a stamping cost model. A full cost model can be developed to estimate should-be costs that are highly credible and more accurate than those developed using just part attributes.
In this post and the previous blog post, we’ve described four methods for building should-be cost models using data that in most cases is already being obtained from suppliers:
- Cost tables
- Linear price models
- Multivariate regression models
- Manufacturing process models
Using these approaches, buyers should be able to answer the “how much will it cost?” question quickly and without becoming overwhelmed with constantly RFQ’ing the supply base. All that’s required is organizing the data you already have.
Want to Learn More?
Register for the upcoming informational webinar “What to Do if Your Purchasing Practices Are Not Providing Should-Be Costs”.