Page 50 - MetalForming-Jan-2018-issue

The Science of Forming
Table 4—Possible-Detection Evaluation Criteria
Detection
Criteria
Error- Proofed Inspection
Gauging
Manual Inspection
Possible Range of Detection Methods
Rank
Absolute Uncertainty
Absolute certainty of non-detection
Cannot detect or is not checked
10
Very Remote
Controls will probably not detect
Control is achieved with indirect or random checks only
9
Remote
Controls have poor chance of detection
Control is achieved with visual inspection only
8
Very Low
Controls have poor chance of detection
Control is achieved with double visual inspection only
7
Low
Controls may detect
Control is achieved with charting methods, such as SPC
6
Moderate
Controls may detect
Control is based on variable gauging after parts have left the station; go/no-go gauging performed on 100 percent of the parts after parts have left the station
5
Moderately High
Controls have a good chance to detect
Error detection in subsequent operations; gauging performed on setup and first-piece check
4
High
Controls have a good chance to detect
Error detection in- station; error detection in subsequent operations by multiple layers of acceptance: supply, select, install, verify; cannot accept discrepant part
3
Very High
Controls almost certain to detect
Error detection in- station (automatic gauging with automatic stop feature); cannot pass discrepant part
2
Almost Certain
Controls certain to detect
Discrepant parts cannot be made because item has been error-proofed by process/product design
1
this assumes that paper being checked is correct. Barcodes are great for track- ing inventory, but only if the barcode isn’t crossed out with a marker without a corrected number written next to it (yes, this did happen).
Table 4 lists possible detection cri- teria and their ratings. The actual cri- teria and ratings should be appropriate for your specific needs and concerns.
Prioritizing Risk
The risk-prioritization number (RPN) is generated by multiplying the ratings for severity (S), occurrence (O) and detection (D):
RPN=S*O*D
RPNs vary between 1 and 1000. The highest RPNs help to define action areas. These efforts might take the form of testing, documenting, tracking, changing design, changing process or changing material—or undertaking many other corrective actions.
Who’s on the FMEA Team?
Maybe you’re a process engineer who has worked in several roles over the years and is pretty familiar with the steps involved in making a good part. Although your instinct may be to build the FMEA yourself...don’t. Con- sider a general contractor who knows how to build a house, but still brings in specialists to handle the plumbing, electrical and drywall. Remember, you aren’t trying to lay out all the steps to be successful—you and your team are trying to document all of the ways that you may be unsuccessful. Get as many disciplines as possible involved: receiv- ing, design, tooling, manufacturing, assembly, reliability, quality and ship- ping all can have an impact.
Helpful Resources
for FMEA Development
SAE International (standards.sae.org/ j1739_200901/) and the Automotive Industry Action Group (www.aiag.org/ store/publications/details?Product- Code=FMEA-4) offer documents to guide metalformers through the devel- opment of FMEAs. Even though both organizations are automotive-focused,
FMEA and other quality tools should be applied at companies of all sizes and industries as an effective way to improve first-pass quality and reduce costs.
Preparing to work on your first FMEA may appear to be a daunting task. Many things can go wrong, including some that you had not even
considered beforehand. But that’s the point. Understanding the risk up front means not having to deal with it during production when you are more likely to apply Band-Aids and temporary fixes. After creating a few FMEAs, the process becomes much easier, and FMEAs will fit right into your standard part-development process. MF
48 MetalForming/January 2018
www.metalformingmagazine.com