3614. A Method For Assessing Mass Data Quality Throughout The Product Development Process
$20.00
SAWE Members get a $200 store credit each year.*
*Store credit coupon available at checkout, click the button in your shopping cart to apply the coupon.
Not applicable to SAWE textbooks and current conference technical papers.
Paper
Abstract
In the automotive industry, developing a product is a few years of journey from concept to a product on wheels. The weight engineers have to predict the mass property from just idea concept in the beginning, and progressively increase its precision over time in line with the product definition’s maturity.
There are many methods of weight estimation with various tools and techniques that are developed by Weight Engineering practitioners. No matter which tool is used, there are always some degrees of uncertainty that might obscure judgement and sometimes result in total lack of confidence in data liability, especially if a making a critical decision is involved. Design progression over time will increase the level of certainty and can help us improve the quality of weight estimates, but it requires a robust data management process.
This paper focuses on how to evaluate the quality of weight estimates and manage it throughout the product development cycle. Combining an uncertainty management technique with a bottom up weight projection method, we are enabled to report the weight of the product in shape of a probability distribution, rather than just one uncertain value. This way of reporting allows us to define the likelihood of occurrence of each failure mode, therefore providing a better visibility for decision making. This also helps to form a discipline for improving the quality of weight estimates throughout time.
The methodology involves estimating the weights in three cases of best, worst and most likely, which in turn forms the average and standard deviation for each low end component. By using the Monte Carlo Simulation technique rolling up weight of those components to the complete product, the average and standard deviation of the vehicle can be obtained, and in case of availability of the failure modes, will be used in evaluating the likelihood of their occurrences. The standard deviation of the final product may also be used as an indication of data health at each stage of the development. The magnitude of acceptable standard deviation has to be continuously reduced throughout the design progression to reflect the increased confidence in our engineering estimates.
