Basic Statistics: Understanding the "Language" of Quality
Accuracy and Precision: Accuracy is the degree to which a measured value approaches the true value (for example, when weighing a 100g weight and getting a result of 99.8g, it is accurate); Precision is the consistency of multiple measurements (for example, when using the same caliper to measure the same part and getting results of 5.00±0.01mm each time, it is precise). Only when the two are combined can the measurement be reliable. For instance, in shooting at a target, accuracy means the bullets are concentrated at the bullseye, and precision means the bullets are concentrated in a small area (regardless of whether it is the bullseye).
Central tendency and degree of dispersion: The average is the "center" of the data (e.g., the average of the dimensions of 10 parts), and the median is the middle value after sorting (more resistant to extreme values, such as in salary distribution); the range is the maximum value minus the minimum value (reflecting the range of fluctuation), and the standard deviation is the average degree to which the data deviates from the average (a more stable dispersion indicator, e.g., in a normal distribution, ±3σ covers 99.7% of the data).
Statistical distributions: The Normal distribution is the "benchmark" for continuous data (such as height and weight), showing a bell-shaped symmetry; the Binomial distribution is applicable to n independent "success/failure" trials (such as the number of heads when flipping a coin); the Hypergeometric distribution represents the number of successes in sampling without replacement (such as the number of defective products when drawing 10 pieces from 50 products); the Chi-square distribution is used to test the independence of categorical data (such as the difference in defect rates among different production lines).
Quality management system: from "system" to "culture"
The essence of quality management is to ensure consistency through systematic methods, and the core system revolves around "customer orientation" and "continuous improvement":
ISO 9001: An internationally recognized quality management standard centered around the "PDCA Cycle" (Plan - Do - Check - Act), covering the entire process from suppliers to customers. Its value lies in "making customers believe in the stability of your quality" - for example, customers can trust the products of enterprises that have passed ISO certification without repeated audits.
TQC and TQM: TQC (Total Quality Control) emphasizes "full participation" (front-line employees form QC teams to solve on-site problems); TQM (Total Quality Management) is an upgraded version that binds quality to corporate strategy and incorporates "customer orientation" (VOC, i.e., customer needs) and "continuous improvement" (such as Toyota's lean production).
QFD (Quality Function Deployment): A tool for converting customer requirements (VOC) into technical requirements (VOE). For example, if customers demand "long battery life for mobile phones", QFD will break it down into "battery capacity ≥ 5000mAh" and "processor power consumption ≤ 5W" to ensure that customer expectations are met at the design stage.
Process control: "Tame" variation with tools
The goal of process control is to ensure that production is "stable and meets specifications", and the core tools focus on "identifying variations and solving problems".
Control Chart: The "barometer" of process stability. By plotting the time series of data and combining the upper and lower control limits (UCL/LCL, ±3σ) and the center line (CL, average), the type of variation can be determined: Variations beyond the limits are due to special causes (such as equipment failures) and need to be dealt with immediately; those within the limits are due to common causes (such as minor material differences) and require systematic improvement.
Histogram: It shows the distribution pattern of data. If it presents a bimodal shape, it may be due to the differences between two pieces of equipment. If it is left - skewed, it may be because large - sized products were removed during the inspection.
Scatter Diagram: Analyze the correlation between variables — for example, the scatter points of "injection molding temperature" and "shrinkage rate" show a linear upward trend, indicating that the higher the temperature, the more serious the shrinkage, and the parameters need to be adjusted.
Pareto: Use the "80/20 rule" to find the main problems. For example, "surface scratches" account for 60% of the defects. Prioritizing the solution to this problem can quickly reduce the defect rate.
Cause-and-effect diagram (fishbone diagram): Find the root cause from the aspects of "Man, Machine, Material, Method, Environment, Measurement" (5M1E). For example, "product breakage" may be caused by "too low heat treatment temperature" (Method) or "insufficient material strength" (Material).
Measurement system: The cornerstone of reliable data
The credibility of measurement data directly affects quality judgment. MSA (Measurement System Analysis) is the core tool for evaluating measurement systems.
Repeatability: The consistency when the same person measures the same part using the same equipment (for example, when using a caliper to measure the same part, the results are all 5.00 mm).
Reproducibility: The consistency when different people measure the same part using the same equipment (for example, if Operator A measures 5.00 mm and Operator B measures 5.05 mm, it indicates that the SOPs are not unified).
EV (Equipment Variation): The error of the equipment itself (e.g., the accuracy of the caliper is ±0.01mm).
GR&R (Repeatability and Reproducibility): A comprehensive variation indicator, with requirements of ≤10% (acceptable), 10% - 30% (needs improvement), and
Failure management: From "putting out fires" to "preventing fires"
The core of failure management is "prevention first", and the key tool is FMEA (Failure Mode and Effects Analysis).
Failure mode: Possible problems (e.g., "The seat belt buckle fails to pop out").
Failure cause: The root of the problem (e.g., "Insufficient strength of the locking spring").
Failure effect: Injury to customers (e.g., "unable to protect personnel safety").
RPN (Risk Priority Number): Severity × Occurrence × Detectability. For example, the RPN of "latch failure" = 9 (Severe) × 3 (Occurrence) × 8 (Difficult to detect) = 216, which belongs to high risk and requires immediate improvement (e.g., replacing the spring material).
FA (Faulty Product Analysis) is the "autopsy" after failure. For example, in the case of "a mobile phone battery swelling", through disassembly, it is found that "the charging voltage is too high", and then the charger circuit is improved. Poka - Yoke is a more proactive prevention method. For example, when the sensor detects that a part is placed incorrectly, the equipment automatically shuts down to avoid the production of defective products.
Quality role: The "goalkeeper" throughout the entire process
Quality management requires clear role division to ensure that every step is done right the first time.
IQC (Incoming Quality Control): Intercept defective materials. For example, when inspecting the strength of steel, if it is lower than 100 MPa, the materials will be rejected to prevent them from entering the production line.
IPQC (In-Process Quality Control): Monitor process parameters - for example, check the temperature and pressure of the injection molding production line every hour. If they exceed the SOP range, adjust them immediately.
FQC (Final Quality Control): Ensure the quality of finished products. For example, in the FQC of mobile phones, functions, appearance, and packaging are inspected. Only products that pass the inspection can leave the factory.
QA (Quality Assurance): Establish rules—for example, write the "Incoming Material Inspection Specification", review suppliers, and handle customer complaints.
QE (Quality Engineering): Solve problems - for example, optimize injection molding parameters using DOE (Design of Experiments) to reduce the shrinkage rate from 5% to 1%.
Traceability is the "ID card" of quality. Through batch numbers and timestamps, the "origin" of products can be traced. For example, if there are defects in a certain batch of products, the raw material suppliers and production equipment can be identified, and the products can be quickly recalled and improved.
Continuous improvement: The "perpetual motion machine" of quality
Continuous improvement is the soul of quality management. The core indicators and methods focus on "better, faster, and more cost - effective":
FPY (First Pass Yield): The ability to do it right the first time. For example, if 85 out of 100 products pass the first inspection, FPY = 85%.
Rolled Yield: The proportion of products that go through the entire process without rework. For example, if the pass rates of three processes are 95%, 90%, and 98% respectively, the rolled yield is approximately 83.8%.
PPM (Parts Per Million Defective): An indicator of high quality—for example, if there are 50 defective items out of 1 million items, PPM = 50 (6 Sigma requires ≤ 3.4).
CPI (Continuous Process Improvement) refers to specific actions. For example, use SMED to reduce the changeover time from 2 hours to 20 minutes and reduce the waste caused by machine downtime; use 5S to organize the work site and reduce operational errors. The prevention cost is far lower than the correction cost (prevention: appraisal: internal failure: external failure = 1:2:4:8). For instance, an investment of 10,000 yuan in the mistake-proofing method can avoid a subsequent rework cost of 100,000 yuan.
The essence of quality management is to "meet customer needs using systematic methods" - from understanding variation through statistics, to ensuring consistency with a system, then solving problems with tools, and ultimately achieving "getting it right the first time and every time" through continuous improvement. Behind all the terminologies lies the simple logic of "putting the customer at the center".
APQP (Advanced Production Quality Planning)
APQP is the first of the five core tools in the IATF16949 automotive industry quality system. It focuses on quality prevention in the early stage of product development. Through the five major stages of "planning → product design → process design → trial production → mass production", quality requirements are incorporated into each step. For example, when an automotive parts enterprise develops seats for a new vehicle model, APQP will use DFMEA (Design Failure Mode Analysis) to identify the risk of "seat adjustment malfunction" during the product design stage - the cause may be insufficient motor torque. In the process design stage, the assembly process is optimized, and a "torque detection" process is added. In the trial production stage, the process stability is verified to ensure that the scratch rate is reduced from 10% to 0 during mass production. Its core value is to "solve problems in the budding stage and avoid more costly rectifications at the back - end of mass production".
SPC (Statistical Process Control)
SPC is a tool for monitoring process variation using statistical methods. Its core is to distinguish between "common causes" (random and inevitable minor variations, such as minor fluctuations in materials) and "special causes" (abnormal and eliminable variations, such as equipment failures). Commonly used tools are control charts (e.g., the X - R chart monitors the mean and range of dimensions, and the P chart monitors the non - conforming product rate). For example, when an electronics factory produces resistors, SPC will collect resistor value data in real - time. If the data points of a certain batch exceed the control limits (e.g., the mean jumps from 100Ω to 120Ω), it indicates the existence of a special cause (e.g., an error in the raw material batch), and the production must be stopped immediately for adjustment to avoid batch non - conformities. The goal of SPC is to "keep the process stable and prevent rather than conduct post - event inspections".
BOM (Bill of Materials)
The BOM is the digital DNA of the product structure, presenting the relationship from the finished product to the bottom - level components in a hierarchical structure (e.g., mobile phone → motherboard → CPU → heat sink). It is the core data of the ERP system, running through the entire processes of procurement (calculating material requirements), production (preparing materials and scheduling production), cost (calculating the cost per unit), and maintenance (finding spare parts). For example, if the "camera module" is omitted in the BOM of a mobile phone manufacturer, it may lead to material shortages and production shutdowns, or customers may complain because the finished products lack cameras. If there are errors in the BOM hierarchy (e.g., classifying the "battery" under the "motherboard"), it will cause errors in the MRP (Material Requirements Planning) calculation and result in inventory overstock.
P.O. (Purchasing Order, Purchase Order)Purchasing OrderPurchase Order Purchase Order P.O. (Purchasing Order)
A P.O. is an official purchasing instruction given by the buyer to the seller and is legally binding. Its contents include the name of the materials, specifications, quantity, price, delivery date, payment terms, and quality requirements. For example, when a manufacturing enterprise purchases steel, the P.O. will clearly state "material: Q235, specification: Φ50mm, quantity: 10 tons, delivery date: June 10, 2024, payment 30 days after the goods arrive". The seller is required to strictly fulfill the contract: if the delivery is delayed, the buyer has the right to claim compensation; if the quality does not meet the requirements (such as the hardness of the steel not reaching the standard), the buyer can reject the goods. A P.O. is the core document to avoid trade disputes and ensures clear rights and responsibilities for both the buyer and the seller.
CAR (Corrective Action Request)
CAR is a rectification instruction for existing problems, and its sources include customer complaints, internal audits, and unqualified inspections. The process needs to be closed-loop: problem description → cause analysis (5Why method) → corrective measures → effect verification. For example, if the customer reports that "the appearance of a certain batch of products is scratched", CAR will require the packaging group to analyze the cause - the original packaging foam is only 1 cm thick and cannot buffer the transportation impact; the corrective measure is to replace it with 2-cm-thick foam; during verification, 3 consecutive batches of products need to be tracked to confirm that the scratching rate has decreased from 10% to 0 before the CAR can be closed. Its core is to "solve problems at the root and avoid recurrence".
PCB (Printed Circuit Board)
PCB is the physical framework of electronic devices. Based on an insulating substrate, conductive circuits are formed by etching copper foil to carry components such as resistors, capacitors, and chips. It is classified into single-sided boards (e.g., toy remote controls), double-sided boards (e.g., routers), and multi-layer boards (e.g., mobile phone motherboards, which can have more than X layers) according to the number of layers. Its quality directly determines the reliability of electronic devices: if the PCB circuit is open, the mobile phone will have no signal; if the circuit is short-circuited, the battery may catch fire. Therefore, in PCB production, the circuit precision (e.g., line width ±0.05mm) and soldering quality (no cold solder joints or bridging) need to be strictly controlled.
COP (Company Operation Procedure/Customer Oriented Process)
COP has two core meanings:
1. Company operating procedures: Internal standardized processes (such as onboarding approval and reimbursement processes) ensure consistent cross - departmental collaboration.
2. Customer-oriented processes (terminology in IATF16949): Focus on "key processes to meet customer needs", such as order receipt, product delivery, and customer complaint handling.
For example, the COP (Order Processing) of automobile manufacturers requires accurate identification of customer needs (such as vehicle model configuration and delivery time) and coordination with the production and procurement departments for timely delivery. If the order information is incorrect, it will lead to production mismatches and cause customer loss. The core of COP is "centering around the customer and transforming needs into process outputs".
PE (Production Engineering, production engineering)
PE is the "designer and optimizer of the production process". Its core responsibilities include: designing the production line layout (such as using U-shaped lines to reduce material handling), formulating operation standards (SOP), selecting production equipment (such as the model of the pick-and-place machine), and solving production technology problems (such as false soldering in welding). For example, when a new factory introduces a mobile phone assembly line, PE will simulate the process: arrange "component placement → welding → testing → assembly" in the order of the process to reduce the material handling distance; at the same time, design tooling fixtures to fix the mobile phone case to avoid misalignment during assembly. The goal of PE is to make the production process "efficient, stable, and low-cost".
ECO/N (Engineering Change Order/Notice, Engineering Change Instruction/Notice)
ECO/N is an official document for product or process design changes. The reasons include customer requirement changes, design optimization, material substitution, and regulatory requirements. The process needs to go through "change application → cross - departmental review (technology/quality/production/purchasing) → approval → implementation → verification". For example, due to the increase in the price of plastics, a home appliance enterprise changes the air - conditioner shell from ABS to PP. The ECO needs to clarify: the content of the change (material substitution), the scope of influence (purchasing changes suppliers, production adjusts the injection - molding process), and the implementation time (starting from the next batch). After the change, the quality department needs to verify the strength of the shell made of PP material to avoid cracking problems caused by the material change.
QA (Quality Assurance)
QA is the "guardian of the quality system". Different from QC's "post - inspection", QA focuses on process prevention: by establishing a quality system (such as ISO9001), auditing the system operation (internal audit), and verifying the process effectiveness (process validation), it ensures that the product quality is "predictable and repeatable". For example, the QA in a food enterprise will audit the hygiene of the production workshop (whether employees wear gloves), the raw material inspection records (melamine detection of milk powder), and the finished product traceability system (production batch of each batch of products). If the system operation gets out of control, it will lead to food safety accidents.
EMS (Environment Management System)
EMS is the environmental management framework of the ISO14001 standard, focusing on the environmental impacts (such as wastewater, waste gas, and noise) throughout the product lifecycle. The process includes: identifying environmental factors (such as "organic solvent volatilization" in chemical enterprises), evaluating risks (such as whether the waste gas exceeds the standard), formulating control measures (installing activated carbon adsorption devices), and monitoring performance (measuring the waste gas concentration monthly). For example, the EMS of a certain factory will control the "pH value of wastewater" between 6 and 9 to avoid polluting rivers. The core of EMS is "environmental compliance + continuous improvement".
QC (Quality Control)
QC is the "last line of defense for product quality", focusing on post - event inspection. It ensures that products meet the standards through sampling or full inspection. It is classified by process as follows:
- IQC (Incoming Quality Control, verifying the qualification of materials);
- IPQC (In-Process Quality Control, monitoring the stability of production processes);
- FQC (Finished Product Inspection, verification before warehousing);
- OQC (Outgoing Quality Control, confirmation before shipment).
For example, in a clothing enterprise, the QC has different tasks. The IQC checks the color fastness of fabrics (to avoid color fading after washing), the IPQC conducts patrol inspections on the sewing process (to avoid skipped stitches), and the FQC checks the dimensions of the finished products (the collar width is ±1 cm). The goal of QC is "to prevent substandard products from flowing into the next process or reaching the customers".
FQC (Final/Finished Quality Control, finished product/completed quality control)
FQC is "the last checkpoint before finished products are put into storage", focusing on comprehensive verification of finished products. The content includes appearance (no scratches on the mobile phone screen), function (the air conditioner can cool), performance (the battery can last for 12 hours), and identification (correct model/batch number). For example, the FQC of home appliance enterprises will conduct a "power-on test" on each refrigerator: run it for 2 hours, check the cooling temperature (0 - 5℃), noise (≤40dB), and the sealing of the door seal (no air leakage). After passing the test, a "qualified inspection" label will be affixed. Omissions in FQC will directly lead to unqualified products flowing to customers.
QE (Quality Engineering)
QE is the "solver of quality and technology". Its core responsibilities include: formulating inspection standards (product dimensional tolerances), analyzing quality problems (failures from customer complaints), promoting continuous improvement (6Sigma projects), and introducing quality tools (FMEA, MSA). For example, the QE of an electronics factory uses the 5Why method to analyze the problem of "swollen mobile phone batteries": swollen battery → high battery pressure → electrolyte leakage → failure of the sealant → insufficient viscosity of the glue → wrong supplier batch. The final measure is to replace the glue supplier and add glue viscosity detection in IQC - to solve the problem at the root.
FRACAS (Failure Report Analyse and Corrective Action System)
FRACAS is a closed-loop problem management system, and the process is as follows: Fault reporting (recording the phenomenon/time/batch) → Fault analysis (using FMEA/5Why to find the root cause) → Corrective actions (replacing parts/optimizing the process) → Effect verification (tracking subsequent batches) → Experience feedback (updating FMEA/SOP). For example, in the automotive industry, FRACAS records the "engine abnormal noise" fault: The cause is that the hardness of the bearing steel fails to meet the standard, the measure is to replace the bearing supplier. After verifying 3 batches, the abnormal noise rate drops from 5% to 0. Finally, "bearing hardness testing" is added to the IQC standard. Its core is to "trace the root cause of every fault".
QM (Quality Management/Quality Manual)
QM has two meanings:
1. Quality management: The whole - process activities of an organization to achieve quality goals, including quality policy (e.g., "Customer satisfaction, continuous improvement"), objectives (e.g., the pass rate of finished products is 99.8%), and system (ISO9001);
2. Quality Manual: The top-level document of the system, which describes the quality policy, system structure, and responsibilities (e.g., the R & D department is responsible for design review, and the production department is responsible for process control).
For example, the quality manual of a certain machinery enterprise will clearly state that "inspection records of each batch of products shall be kept for 3 years" as the basis for traceability. If there is a problem with the products, the responsible department can be quickly located.
IE (Industrial Engineering)
IE is an "expert in optimizing efficiency and cost". Its core tools include: 5S (Sort/Set in order/Shine/Standardize/Sustain), Lean Production (eliminating waste), Therblig Analysis (reducing ineffective movements), and Time Study (establishing standard working hours). For example, the IE department of a clothing factory found through Therblig Analysis that workers had to turn around to get needles and thread when sewing buttons (taking 2 seconds per piece). So they placed the needle and thread box beside the sewing machine, and a single production line produced 50 more pieces per day. At the same time, they used Time Study to establish a standard working hour of "sewing 10 buttons in 1 minute" as the basis for production capacity assessment. The goal of IE is to "produce the most value with the least resources".
QMS (Quality Management System)
QMS is a "system framework for an organization to achieve its quality objectives". It is established in accordance with standards such as ISO 9001 and IATF 16949, and includes documents (quality manual/procedure documents/SOP), processes (procurement/inspection/complaint handling), and resources (personnel/equipment/funds). For example, the QMS in the manufacturing industry covers the entire process of "customer order → raw material procurement → production → delivery → feedback": order review to confirm requirements, IQC to verify materials, IPQC to monitor production, OQC to ensure delivery, and CAR to handle complaints. The core of QMS is "standardization to avoid human errors".
IPQC (In-process Quality Control, Process (Operation) Quality Control)
IPQC is the "quality sentinel in the production process". Through regular inspections (once every 2 hours), it verifies the following: process parameters (injection molding temperature of 180℃), operation methods (workers operate according to SOP), and semi - finished product quality (no false soldering on the circuit board). For example, the IPQC in an electronics factory will check the "pick - up accuracy" (whether the chips are accurately picked up) and "placement pressure" (whether the chips are crushed) of the placement machine in the SMT process. If placement deviation is found, the parameters need to be adjusted immediately to avoid batch errors. The value of IPQC is to "detect process variations in a timely manner and prevent batch non - conformities".
QP (Quality Plan/Planning/Procedure)
QP contains three layers of content:
1. Quality plan: Quality arrangements for specific products/projects (such as design review nodes and inspection points for new products);
2. Quality planning: Set quality objectives and paths (e.g., "Reduce the complaint rate by 50% in 2024, and SPC needs to be introduced").
3. Quality procedures: Standardized operating processes (e.g., non-conforming product handling procedures, customer complaint processes).
For example, the quality plan for a new product will clearly state that "the design review will be completed in the 3rd month, trial production will begin in the 6th month, and mass production will start in the 9th month", and stipulate that "PFMEA analysis should be conducted during trial production" to ensure that the quality of the new product is controllable.
IQC (Incoming Quality Control, Incoming Material Quality Control)Incoming Material Quality Control Incoming Quality Control
IQC is the "first line of defense for materials entering production", verifying whether the materials meet the procurement requirements: appearance (steel without rust), dimensions (screw diameter Φ5mm ± 0.02mm), performance (plastic melting point ≥ 150°C), material (composition of stainless steel 304). For example, the IQC in an automobile factory will conduct the following inspections on "automobile tires": appearance inspection (no cracks), dimension inspection (tread width ± 10mm), performance test (abrasion resistance index ≥ 400) - the tires will be put into storage after passing the inspections. If the IQC misses something, it will cause "tires with non - compliant dimensions" to enter the production process, resulting in assembly difficulties or potential safety hazards.
QR (Quality Record)
QR stands for "evidence chain of quality activities", including inspection reports, audit reports, customer complaint records, and production logs. Its core value lies in traceability: if there is a problem with the product, "what happened, who is responsible, and how to handle it" can be restored through QR. For example, the QR of a certain milk powder enterprise records "the raw material batch of each batch of milk powder (dairy farm number), production time (May 1, 2024), and inspector (Zhang San)". If there are foreign objects in the milk powder, it can be traced back to "the raw materials come from Dairy Farm A and the filling machine was not cleaned during production", so as to quickly locate the root cause.
LCA (Life Cycle Analysis, life cycle analysis)
LCA is an "environmental assessment tool for the entire product lifecycle" that covers four stages: "raw material extraction → production → use → disposal" and assesses energy consumption, water consumption, and pollutant emissions (such as CO₂ and heavy metals). For example, the LCA analysis of a mobile phone manufacturer: in the raw material stage (lithium ore mining consumes a large amount of water), in the production stage (chip manufacturing has high energy consumption), in the use stage (battery charging consumes electricity), and in the disposal stage (it is difficult to recycle the plastic casing). Based on LCA, manufacturers can make optimizations: use recycled plastics for the casing (reduce mining) and improve battery life (reduce charging) - to reduce the environmental footprint.
TQM/C (Total Quality Management/Control, Comprehensive Quality Management/Control)
TQM stands for "Total Quality Management concept with full participation", and its core is the "Three Totals":
All staff: Everyone from senior management to front-line employees is involved (e.g., workers submit improvement suggestions).
Whole process: Every step from design to after-sales service (such as design review, production inspection, customer feedback);
All elements: people, machines, materials, methods, environment, and measurement (such as equipment maintenance and raw material inspection).
For example, a home appliance enterprise implements TQM with the following requirements: Workers are required to propose "changing the tooling from iron to plastic (to reduce weight)", the department conducts a "6Sigma project to reduce the noise of air conditioners", and senior management sets the goal of "reducing the complaint rate by 50%". The goal of TQM is "continuous improvement to satisfy customers".
ME (Manufacturing Engineering, )
ME is the "designer of manufacturing processes". The core responsibilities include: formulating manufacturing processes (such as "stamping → welding → painting" for automotive parts), designing tooling fixtures (jigs for fixing parts), optimizing process parameters (e.g., welding current of 100A and time of 2 seconds), and solving manufacturing problems (such as part deformation). For example, the ME in an automobile factory designs a "positioning fixture" in the "door welding" process to fix the car door and door frame, ensuring that the welding deviation is ≤ 0.5mm; at the same time, optimizes the welding sequence (from the middle to both sides) to reduce deformation. The goal of ME is to make the manufacturing process "stable, efficient, and meet the design requirements".
PPAP (Production Parts Approval Procedure)
PPAP is one of the five major tools of IATF16949. It is a "document package for suppliers to prove their capabilities to customers". The documents to be submitted include: PSW (Part Submission Warrant), samples, inspection reports, DFMEA/PFMEA, control plans, and process parameter records. For example, if an automotive supplier provides "engine blocks", PPAP needs to prove that 100 blocks produced on the mass - production line meet the requirements in terms of dimensions (cylinder diameter ±0.01mm) and performance (compressive strength ≥200MPa). At the same time, the supplier should submit PFMEA to prove that the risk of "casting porosity" has been identified and the measure of "increasing flaw detection" has been taken. Only after the customer's approval can the supplier start mass production. The core of PPAP is to "prove the stable production capacity with samples under production conditions".
MRB (Material Research Board, Material Research Meeting (Non-conforming Material Disposal))
MRB is the "Review and Decision - Making Body for Non - conforming Materials", which is composed of the procurement, quality, production, and engineering departments. Its responsibility is to determine the disposal methods for non - conforming materials.
Concession acceptance (does not affect performance, such as minor scratches on plastic parts);
Rework/Repair (e.g., if the screw is too long, cut it to the standard size);
Return (the material does not match and it cannot be used);
Scrapped (damaged and has no repair value).
For example, the MRB of an electronics factory is dealing with the issue of "a 10% deviation in capacitor capacitance" (the standard is ±5%). If the capacitors are to be used in ordinary circuits, they can be accepted with concessions; if they are to be used in high-precision circuits, they should be returned. The MRB's decision needs to balance "quality risk" and "cost efficiency".
WI (Work Instructions)
WI is the "operation bible for front-line employees" and the most detailed standard document. Its contents include: operation steps (take materials → install fixtures → set parameters → start equipment → check results), tools/materials (use a Φ3mm screwdriver), parameter requirements (temperature 200℃, time 3 seconds), and precautions (wear heat-insulating gloves).