PPM analysis – statistical risk assessment for 0-km and warranty returns

In case of quality problems that are not one single occurrence, we may be required by client or organization to prepare a risk assessment related to potential non-conforming products population.

One of such assessments is the PPM analysis, which takes into account the dependence of affected parts quantity in relation to a million manufactured parts (PPM – parts per million).

Its unquestionable advantage is the fact that it can be used to carry out risk assessment both for units reported from the customer’s production plants (0-km) and from the warranty. By the organization, it’s most often applied in the following situations:

  • inability to use process data that could help define the affected population
  • drawing values are changed during the production process without the possibility of their reconstruction. For example shaft diameter on which are pressed the bearings.

In order to use it, you do not need any additional software (as in the case of the Weibull analysis, for example) – a standard Excel form is enough.

PPM analysis – assumptions

Each risk analysis is based on assumptions that must be considered in its preparation. We have to remember that it should be included in the documentation presented to the client or within the organization.

In the case of PPM analysis, such assumptions include:

  • Start date defining from which we are able to identify the beginning of the problem. If this is not possible, the start date will be equal with Start of Production (SoP) of particular project.
  • Closing date defining that corresponds to the date of permanent corrective action (PCA) introduction.

These two dates are our cut-off points that define the time range which will be considered during the preparation of the risk assessment.

  • It can be applied for minimum two returned units. Complaint  for assigned failure mode which we deal with only for one returned unit is presented as “one single occurrence”).
  • It requires an appropriate update when the organization receives new parts for analysis.
  • It is considered on a monthly basis. For all months including the risk assessment, we should receive information from the logistics department or directly from customer portals about  number of parts shipped. It should be both for months for which no defective parts were found and for those where such parts were reported by the customer as non-conform.

PPM analysis – form overview

In above risk assessment, we will use four columns that contain the following names:

  • “Month”
  • “Returned units” assigned to individual months
  • “Shipped units” shipped in individual months
  • “PPMs counted” that are only calculated for months where no problem was identified. The formula for this factor is as follows:

Discussed form is illustrated in the below example. As we see start date is July and the end date is March. This is the time period which will cover nine months in total.

PPM analysis - form overview

Table. PPM analysis – calculations example

The number of returned units is considered for the following months: July, October, December, January, February, March – 42 in total. For the remaining months, we will calculate PPM based on the results from the “PPMs counted” column.

In the “Shipped units” column, we insert the monthly amounts obtained from the logistics department or customer portals. On the other hand, the last column “PPMs counted” is used to calculate the values for the months in which defective units appeared.

The most important element of the PPM analysis is to include the highest value in the next step from the “PPMs counted” column. It has to be done as the “worst-case scenario”.

Analysis final stage is PPMs calculation for months: August, September and November using the following formula:

By this, we can calculate the final number of parts that will be considered in the risk assessment. In the above case, the number is equal 82.58 and must be rounded to 83.

In the documents provided to the customer (OEM), it’s also worth including the statement, that the potential number of affected units is equal 83, of which currently have been identified 42. Potentially 41 parts remain in the warranty field or at the customer’s plant.

Download our free ebook Automotive Warranty Management – Practical Guideline for Suppliers

You can download an automatic, editable Excel form for free on the Free Quality Tools

Document name: PPM analysis – Risk Assessment method for 0-km claims and warranty returns – Excel form

Download for FREE our E-BOOKS

X