Intellectual quality management method, electronic device and computer readable storage medium

Abstract

An intellectual quality management method is disclosed. A heatmap risk interface is created according to the required data and the parameter configuration which are calculated using a time dependent risk priority number (RPN) equation. An intellectual audit scheduling algorithm is defined via the heatmap risk interface to automatically generate at least one audit plan. An audit program corresponding to the audit plan is performed and a plurality of problem points are selected. Intellectual root cause category recommendation is performed to the questions points. intellectual corrective actions and preventive action recommendations are performed to the problem points according to the intellectual root cause category recommendation to obtain at least one optimum corrective action and at least one preventive action. Corrective actions are performed to each audit unit according to the corrective action to solve the problem points and prevention actions are performed to each audit unit according to the preventive action.

What is claimed is: 
     
         1 . An intellectual quality management device, comprising:
 at least one processor;   a storage unit; and   one or more programs that are stored in the storage unit and executed by the at least one processor, the one or more programs comprising instructions is configured for the at least one processor to process:   calculating required data and a parameter configuration according to audit data using a time dependent risk priority number (RPN) equation;   creating a heatmap risk interface according to the required data and the parameter configuration;   defining an intellectual audit scheduling algorithm via the heatmap risk interface to automatically generate at least one audit plan;   informing an audit unit to check an audit time and an audit team and performing an audit program corresponding to the audit plan and selecting a plurality of problem points;   performing intellectual root cause category recommendation to the questions points;   performing data cleaning and preprocessing raw data of the problem points, wherein the data cleaning and preprocessing comprises data labeling, Chinese-English word segmentation modeling, and quality key word bag creating;   performing intellectual corrective actions and preventive action recommendations to the problem points according to the intellectual root cause category recommendation to obtain at least one optimum corrective action and at least one preventive action; and   performing corrective actions to each audit unit according to the corrective action to solve the problem points and performing prevention actions to each audit unit according to the preventive action.   
     
     
         2 . The device of  claim 1 , defining the intellectual audit scheduling algorithm via the heatmap risk interface to automatically generate the audit plan further comprises instructions for:
 selecting N audit units and M problem points to be audited at a current time period;   calculating time dependent RPN values of K problem points;   calculating time dependent RPN values of each of the N audit unit;   selecting a plurality of audit units whose RPN value rank before X-th;   determining whether the current time period is a first period of a circulation period;   if the current time period is not the first period of the circulation period, determining whether there is any duplicated audit unit in selected audit units;   if there is at least one duplicated audit unit in the selected audit units, removing the duplicated audit unit;   if the current time period is the first period of the circulation period or there is no duplicated audit unit, indicating that the selection of the N audit units is complete and selecting the problem points of the former Y big RPN values and the problem points of the second Z big RPN values, wherein Y+Z=M;   determining whether the current time period is the first period of the circulation period;   if the current time period is not the first period of a circulation period, determining whether there is any duplicated problem point in the selected problem points;   if there is at least one duplicated problem point in the selected problem points, removing the duplicated problem point;   as the problem points of the former Y big RPN values are selected, or if the current time period is the first period of the circulation period, or if there is no duplicated problem point, indicating that the selection of the M problem points is complete and determining whether any one of quality incident is detected within a preset time;   if there is at least one quality incident is detected, configuring the selected problem points as multiple cross-auditing unit projects;   if there is no quality incident is detected, assigning L problem points whose RPN value exceeds a preset value to each audit unit; and   proceeding to an audit of a next time period.   
     
     
         3 . The device of  claim 1 , wherein the audit program comprises creation of new quality cases and description of filling and searching root causes. 
     
     
         4 . The device of  claim 1 , wherein the time dependent RPN equation is represented as: 
       
         
           
             
               
                 
                   RPN 
                   question 
                 
                 = 
                 
                   
                     
                       Weight1 
                       × 
                       
                         Security 
                         ( 
                         S 
                         ) 
                       
                       × 
                       
                         Occurrence 
                         ( 
                         O 
                         ) 
                       
                       × 
                       
                         Detection 
                         ( 
                         D 
                         ) 
                       
                     
                     + 
                     
                       Weight2 
                       × 
                       
                         
                           PFEMA 
                           ⁢ 
                           score 
                         
                         ( 
                         p 
                         ) 
                       
                     
                   
                   = 
                   
                     
                       
                         W 
                         1 
                       
                       ⁢ 
                       
                         
                           ∑ 
                           i 
                         
                         
                           
                             ∑ 
                             k 
                           
                           
                             
                               ∑ 
                               m 
                             
                             
                               
                                 S 
                                 
                                   i 
                                   , 
                                   j 
                                   , 
                                   k 
                                   , 
                                   m 
                                 
                               
                               × 
                               
                                 O 
                                 
                                   i 
                                   , 
                                   j 
                                   , 
                                   k 
                                   , 
                                   m 
                                 
                               
                               × 
                               
                                 D 
                                 
                                   i 
                                   , 
                                   j 
                                   , 
                                   k 
                                   , 
                                   m 
                                 
                               
                             
                           
                         
                       
                     
                     + 
                     
                       
                         W 
                         2 
                       
                       × 
                       
                         p 
                         j 
                       
                     
                   
                 
               
               , 
             
           
         
         wherein “Security” indicates severity, “Occurrence” indicates incidence, “Detection” indication detectability, “i” indicates quarterly, “N” indicates questions, and “k” indicates event types. 
       
     
     
         5 . The device of  claim 1 , further comprising one or more programs for:
 connecting a database, PostgreSQL, and front-end network application frameworks like html, css and Vue.js via an application programing interface (API) to construct the heatmap risk interface.   
     
     
         6 . The device of  claim 1 , the at least one processor is further to process:
 performing model training and prediction model creation to the raw data of the problem points;   wherein:   the model training comprises training model updating and ranking calculation for corrective actions (CA) and preventive actions (PA); and   the prediction model creation comprises main cause prediction of quality problems and recommendation of improvable and preventive actions.   
     
     
         7 . The device of  claim 6 , wherein the ranking calculation for the corrective actions and the preventive actions further comprises one or more programs for:
 discovering the most similar narrative to the current situation from a database relative to past actions implementation and recommending a corresponding actions;   wherein an evaluation score of the ranking calculation for the corrective actions and the preventive actions is W1+W2, W1 refers to reduce the evaluation score of repeated attributes, while W2 refers to effectiveness of professional scoring.