Predictive indicator model
A predictive indicator tool for predicting the risk of a manufacturing process producing products having defects. Manufacturing sub-processes are identified for the process and s a series of 12 questions is applied to each sub-process. The questions are generated using Six Sigma statistical techniques. The questions are scored and an associated algorithm is applied to produce a quantitative score is indicative of risk. The quantitative risk score is compared to an alert level.
1. A method for manufacturing an improved medical device, comprising:
providing a suitably programmed computer for calculating a risk score associated with the manufacturing process;
executing upon the suitably programmed computer a predictive indicator tool for analyzing the manufacturing process to determine each sub-process, the predictive indicator tool for each manufacturing sub-process including:
A. 1) inputting the Level of Patient Risk due to potential failure of the medical device,
2) inputting the Level of Regulatory Risk related to the medical device,
3) inputting data related to whether product CTQs are identified and linked to VOC,
4) inputting data related to whether a Product Risk Assessment is complete and a risk mitigation plan is in place,
5) inputting data related to whether CTQs have flowed down to component/RM level with control plans in place for each component/RM,
6) inputting data related to whether Sampling Plans/Control strategies utilize documented statistical rationales,
7) inputting data related to whether a Control Plan Exists and identify KPIVs and controls for each CTQ,
8) inputting data related to whether an Adequate Test Method control is identified for each CTQ,
9) inputting data related to Critical Defect (3 month average) Process Capability in DPM or PpK (complete roll-up),
10) inputting data related to whether there is an Effective Training Plan defined, followed, and up to date,
11) inputting data related to whether manufacturing Methods and other procedures are properly executed (# of human error deviations),
12) inputting data related to whether Process FMEA is complete and adequate,
B. scoring the data inputted for each of the twelve data input areas and determining a risk score for the process by using the following algorithm:
PI=Severity*Detection*Occurrence/23328*100, wherein
Severity=the sum of the scores of Questions 1 and 2,
Occurrence=the sum of the scores of Questions 3, 4, 5, 7, 9, 10, 11, 12
Detection=the sum of the scores of Questions 6 and 8; and
C. outputting the risk score to the user.
2. The method of claim 1 , comprising the additional step of ranking and prioritizing the sub processes.
3. The method of claim 2 , additionally comprising the step of developing a mitigation plan to decrease the risk score.