By definition, the MES (Manufacturing Execution System) must fulfil 11 functions, including Quality Management, which can take many forms, from the most basic to the most complete.

The input data for this function will be the entries made by the operators through the MES interface, which fulfil 2 main functions:

• Ensure a level of product compliance (by capturing measurements)
• Ensure full traceability of processes (who, when, how)

These two types of data entry are specific: they only concern the production order and/or the part being manufactured. Whatever the type, they have a cost, whether material (acquisition of measuring equipment, metrological checks, etc.) or human (data entry time, metrological training, etc.).

In the absence of SPC (Statistical Process Control), the data is only used at the time it is recorded in order to confirm that the dimension is maintained. As well as the traceability in the database, which is only done in the event of an audit. You are therefore depriving yourself of a second analysis after the fact, which allows a global analysis of the process.

Once the measurement has been capitalised, a second exploitation is possible.
What better way than with an SPC system to turn a one-off view of a part into a global view of the process?

From one measure, you can generate different levels of analysis, for different people or departments, at different times and for decisions that will have a different impact.

Conformity of the OF and/or part: guarantee that the part meets the customer’s specifications
• Decrease in scrap costs and customer returns

SPC alarms : process drift before Production produces non-conforming parts
• Reduction in the cost of creating out-of-tolerance parts

Process capability (Cp Cpk) : probability of making good parts
• Reduction of costly control operations

Evolution of capabilities over time, comparison between means of production, correlation
• Optimisation of production resources to focus corrective actions and investments on the key points of the factory

It would be a shame to miss out on an efficient SPC system that will have a real and immediate impact on the ROI of an MES deployment.

Discover our SPC solution

One of the bases (theoretical and practical) of SPC is to understand and reduce sources of process variation known as the “5M’s”: Machine, Manpower, Method, Material, Mother nature (environment). We could add a 6th M (Management).

But what about the 6th M: Measurement?

One could argue that measurement error is not actually changing the product as it is only checking something already manufactured. However, we must consider the 2 primary reasons for measuring the product:

• 1. Checking the product conforms to customer requirements

When a product is close to the upper or lower tolerance limits, measurement variation can cause two costly issues: either a bad part can be passed to the customer (internal or external), or a good part can be failed resulting in scrap or repair.

• 2. Monitoring and control of the manufacturing process

If the measurement system is not consistent, it can result in unnecessary process setting adjustments being made (known as ‘tweaking’) causing the process to become unstable. This is how measurement variation can directly increase product variation.

Introduction to Gage R&R

A measuring device, and more generally a measurement system (also subject to 5M) has variation and provides a measurement uncertainty.

Gauge R&R (repeatability & reproducibility) is a means of assessing the measurement system variation and concluding whether a gauge is suitable for the intended application.

Practically, an R&R study aims to separate the causes of the measurement variation into two main categories – variation due to the measurement system itself (repeatability), or variation due to the user method or process (reproducibility)

A preliminary Gage R&R study should be conducted prior to introducing SPC or process capability analysis to ensure the validity of the results.

Methods

A Gage R&R study is carried out in the following way: several operators (2 or 3) measure a number of parts (5-10) several times (2 or 3). The order of the measurements is randomized, and an attempt is made to select representative parts which cover the entire tolerance interval.

This study provides information on:

• Repeatability: Variation seen when the same operator measures the same part multiple times. This is an indication of the variation due to the measurement equipment.
• Reproducibility: Variation seen between operators when multiple operators measure the same part multiple times. This is an indication of the variation due to the measurement method.
• Interaction (ANOVA / ANAVAR study) between the parts and the operator. This highlights if there is more variation on particular parts (e.g. top or bottom of tolerance)

The result of the study is a calculated indicator where the overall observed variation is expressed as a percentage of the part tolerance.  This is known as Gage R&R%.

We use this value to conclude whether the measurement system is either:

• Acceptable
• Conditionally accepted (under specific conditions)
• Unacceptable (system needs replacing or improving)

Because the acceptance criteria is partly dependent on the product tolerances for which the measurement system will be used, an instrument may be deemed acceptable for measuring one product or characteristic but not acceptable for another.  Therefore, care should be taken when using an R&R study as a qualification process for purchasing new measurement equipment – make sure that the study represents all the intended products for which it will be used!

Additionally, be mindful that the purpose of a conventional R&R study is only examine the repeatability & reproducibility and validate whether the measurement system is suitable for checking and controlling a product.  A more complete assessment of the gauge should include other characteristics such as accuracy and linearity etc.

SPC Vision software provides functions to conduct a full R&R Study (in accordance with ISO TS 16949  Measurement System Analysis standard) and also CMC indicators (developed by CNOMO Standard with the Cpc indicator introduced by Maurice Pillet).

For advanced interaction calculation methods, we use the ANOVA (Analysis of Variance of the MSA) and its variant ANAVAR (Analysis of the Variance).

Ben Allister & Frédéric Henrionnet
CT Infodream

In this article, we will discuss the concept of inertial tolerancing. We will see how this way of tolerancing applies to the characteristics of a part but also to the control of processes using control charts.

Definition of inertial tolerancing

Inertial tolerancing is a concept developed by Maurice Pillet at the University of Savoie. It is based on a simple observation: the current tolerance system is not always adapted to.

In the worst case, it sometimes imposes tolerances that are impossible to maintain in production, or at too high a cost.
Statistical tolerancing allows tolerances to be extended by taking into account the unlikely cases of parts being assembled at the extreme edge of tolerance. It presupposes the correct centring of the production, and the unlikely cases become much more probable at the slightest decentring.

The basic idea is therefore to return to the Taguchi loss function

The cost is minimised when you are on target. The objective is no longer to put the characteristic within a tolerance interval, but as close as possible to the target!

The inertia is calculated as a function of the standard deviation and the deviation from the output target. You can allow for more dispersion if you are well centred.
Similarly, it is possible to accept a decentring of the average if the dispersion is low. This makes it possible to guarantee the final robustness of the finished product whatever the batches of parts used. This is provided that the specified inertias on these batches are respected..

What does inertial tolerancing offer?

Inertial tolerancing therefore offers :

• another way to tolerate with tolerances:

– inertial : 10 0,003 STI
– arithmetic : 10 ±0,01
– statistics : 10 ±0,01 STQ

• other indicators such as :

– I (Inertia)
– Cpi / Ppi (inertial capability)

• but above all a different way of thinking, a new paradigm :

– the aim is to limit the variability around the target (the target is placed in the centre) ;
– we get out of the inconsistency of classical tolerancing. Indeed, the latter explains that sometimes it is necessary to adjust while the products are good (in the sense of conformity). And that sometimes you have to leave the process when the products are not good (still in the sense of conformity) !

Inertial tolerancing for process control

The next step is naturally to move on to inertial process control.
The process control involves the establishment of a control chart, where the inertia of the samples is monitored. This control chart will have 4 zones (green, orange, red and black) each determining a different action to be carried out according to the position of the point.

The sample can also be viewed in a ½ inertial circle. The dispersion is on the ordinate, and the centring on the abscissa. These 2 simultaneous dimensions allow the sample to be perceived at a glance.

Inertial piloting is bound to change habits. This change is deliberate and accepted because « you have to be close to the target » and not « you just have to be within the tolerances ».

However, in order to maintain the visualization of the classic Shewart charts, we at Infodream have decided to offer a triplet of charts:
– inertial steering
– averages
– standard deviations

Inertial tolerancing for industrial process control

The application of all these steps forms what Maurice Pillet calls « Inertial Process Control ».

These steps are defined and exist in our Qualaxy SPC software:
• Definition of the characteristics
• Calculation of an inertial tolerance
• Determination of extended limits and/or inertial limits
• Control with the 4 zone inertial control chart
• Visualisation of the ½ inertial circle sample / sample or on a population
• Histogram analysis with mean excursion segment and dispersion segment allowed.
• Calculation of inertial capability indicators

What are the applications of inertial process control?

• Assembly of parts with worst-case tolerances that are too tights
• Plastic injection moulding
• Multi-cavity

The purpose of this article is to present the concept of « Inertial Process Control » in a general way. All calculation formulas, other statistical notions and steering rules have been deliberately omitted.

Frédéric Henrionnet
CT Infodream

Sources :

• Standard NF XP E 04-008 / UNM 08D-038
• Maurice Pillet’s work on the subject
• Screenshots from Qualaxy SPC software

How and when to make the transition from the project mode of the SPC approach to its daily use? Who should take the baton? These questions arise for all types of project, but we have some answers for you.

Leading an SPC project is above all applying a project methodology, but it is also a continuous improvement approach and above all a change management.

It is a unifying project that brings together the Methods, Quality, Production, Design Office and General Management departments… It is therefore a structuring project for the company.

The success of this project requires involvement, means and resources.

Let’s assume that all the prerequisites have been met and that the project has been successfully completed. Who can be handed over after an SPC project?

SPC as a production tool

We find ourselves in a more general problem of perpetuating good practices in an industrial environment. The transition from project mode (with a project manager, actors and allocated resources) to serial mode is a crucial stage that must not be neglected.

The sine qua non condition for the sustainability of an SPC project is that Production, operators, adjusters and the workshop manager are responsible for the system: the SPC must be considered as a production tool and not as a constraint. Be careful to include production in the deployment project from the beginning of the project phase to facilitate the handover.

Use of an adapted, simple and fast software

By using high-performance software that is well integrated into the existing system, production work will be facilitated (automatic measurements with connected instruments, drawing of the part and dynamic guidance of the operator during the inspection run, real-time alerts in the event of a drift or alarm, etc.).

The appropriation of the software by the operators is essential to the success of the transfer.

The software will make it possible to standardise uses and representations (control charts, statistical indicators, alert rules, etc.).

The acquisition of measurements must be carried out either by connecting the checking fixtures, or automatically by file transfer, or manually by the operators.

Consolidation of the fundamentals

Integrating SPC into the company’s culture means :

• training of all stakeholders in the SPC and the chosen software
• support for your employees in managing change
• explanation of the objectives of the approach
• appropriation of SPC vocabulary (Cp, Cpk, capabilities) as elements for discussion (quality meetings, etc.)
• the appropriation of the SPC approach as a brand image (internal and external)

Promotion of the work of the production teams

• Daily or weekly review of a few minutes in the workshop, motivation of the teams, emulation of the workshops, ideas for improvements, gains obtained, standardisation of the machine park following an event, etc.)
• Presentation of success stories and their actors
• Presentation of maps, histograms, evolutions, type of alarms, Pareto, indicators… (a few minutes per day allow you to be more reactive and efficient in the reporting of information!)

SPC as a business objective

Empowerment of each individual

Each sector is responsible for the whole chain, not just its own area…

From the design stage, integrate your machine capabilities, communicate between departments, which does not mean picking up the phone to deal with a request, but going to the workshop, talking to people, understanding their jobs, their needs, … and acting in the company’s direction.

The design office must take into account the notions of capability but must also ensure that the tolerances indicated on the drawings are indeed « realistic » in relation to the constraints of production.

Between theory and practice…it’s always the same story! We all know this only too well…

« Theory is when you know everything and nothing works. Practice is when everything works and nobody knows why. Here we have combined theory and practice: Nothing works… and nobody knows why!  »

Albert Einstein

Designate a referent

Operators, adjusters and workshop managers must be responsible for it: the SPC is a production tool. Nevertheless, it is necessary to designate and appoint an SPC manager. This can be a person on the shop floor who is involved in the project and who is in a continuous improvement process, « your field leader ». He or she will be your best representative.

Communication

Invest in strong and determined communication:

• External communication towards customers, the SPC as a competitive vector
• Regular audits (internal or external) in the sense of assistance, advice
• Involvement of senior management: communication of objectives and results, integration of « SPC objectives » into objectives and individual interviews,…

Conclusion

These few elements are well known and usual, as for any classic « improvement » project in an industrial environment. Nevertheless, they must be borne in mind from the start of the project and it must be realised that the appropriation of such an approach by the production team may take time and effort, even if the resources allocated are less than during the project phase.

By defining from the beginning of the project or as soon as possible the person who will be in charge of the relay, you will perpetuate your SPC approach.

Frédéric Henrionnet
CT Infodream

In this article, we will briefly discuss some of the commonly held misconceptions about SPC. In every manufacturing sector, there are people who for one reason or another are not convinced by the benefits that SPC can bring. This is usually due to lack of understanding or previous bad experiences of SPC. We’re not going into too much detail on each point so feel free to get in touch to discuss further.

Statistics ? Too complicated !

Not at all, the calculations and application of the statistics is completely transparent to the user and can be managed by software! The software calculates indicators, creates charts and automatically triggers alarms etc.

However, the software tool is just one element in the establishment of a comprehensive approach to SPC. Also required is strong management support, training and a clear framework of methods and processes for SPC. The level of SPC knowledge required by 90% of the actors is quite basic and focuses on the process –  It certainly does not involve calculations and statistics!

Important: you should select a software which can be easily configured to suit your needs (at all levels including Statistics). The software vendor should also be able to provide training, guidance and SPC methodology support for all phases of the project from pilot to full production.

Statistics ? Only applicable to high volume production !

This is still a widely held belief based on traditional application in the automotive industry. However, it is possible use SPC in low volume or small batch applications. By combining characteristics from similar products and processes (families) it is possible to drastically increase the quantity of data on which the statistics can be applied. These methods are actually much more effective than simple individual product control as they allow you to analyse and understand the ‘global process’.

Example : A typical aerospace part (low volume) is “full of holes”. We will create a larger sample size by grouping the measured diameters according to the tool type, diameter and thickness of the material concerned. This method provides a large amount of data and the analysis is extremely useful.

SPC ? This is a customer requirement, not a general business need !

This mindset is a real shame! This cost of quality control for a company is high. Inspection time, measurement instruments, training of operators and inspectors, calibration, document generation and management etc. All of this has to be done anyway to validate the product so why not use the collected data to add value to the process?

Creating customer conformance reports and providing the required statistics can be done automatically using software. However, of much greater value is how the data analysis can be used to help you improve the manufacturing process by reducing waste. As well as process capability analysis and reporting, SPC software can provide real time feedback on the process performance and is a very effective tool for process control.

Methodology point :

A measurement taken in a workshop is useful from 2 points.

1. Producer (manufacturer) with control charts and alarms – used for process control
2. Final customer with histograms, capability analysis and conformance report.

If the process control charts are used correctly (the process is under control), the final results (customer view) will be correct and consistent!

SPC ? Only for Machining Processes ?

SPC methods can and should be applied to all processes on which we can act and adjust. Machining is historically popular application for SPC but variables exist on all processes:

• Temperature, pressure, Vacuum, speed
• Waiting time and temperature for composites

Methodology point : Remember, as well as providing process control rules and guidance, a SPC system keeps a log (historical) of production. This notion is of course applicable to any manufacturing – automated, manual, high or low volume.

SPC ? Only for large companies !

In any production environment, SPC will provide immediate knowledge and decision making rules to help control the process.

An SPC project is a transverse project where teams from Methods, Quality, Production, Executive Management, and Design Office all work together. These functions cannot always be clearly divided because in smaller companies one person may be responsible for several areas (methods, quality and production, for example), but the scope of the project remains the same: we want to control the entire manufacturing process and learn more about the process to develop future products more efficiently.

SPC ? Only for short term problem solving !

No, do not stop using SPC soon as a problem is solved!

Phase 1 of a project may well be to use SPC to help determine the root cause and corrective actions to solve a quality problem. SPC control charts and statistical analysis can then be used to validate the solution and quantify the improvement. However, this is not the end!

We can move to phase 2 to ensure ongoing control of the process. The difference in this phase is that we are monitoring the key variables (key characteristics) which have the biggest effect on the process and which should have been identified in stage 1. We need to monitor the changes and drift of these variables carefully.  It is at this point where it may be possible to use SPC to reduce inspection activities (and costs) because SPC will give us quantifiable evidence and confidence to do this.

Methodology point : we often speak of drift and variation in a negative way but it’s the same if the process improves and we don’t detect it! If we cannot identify what enabled an improvement, how can we sustain it?

Conclusion

A well run and well supported SPC project always makes progress. Depending on the application, one can easily obtain a quantifiable ROI (by lower scrap and rework), a better understanding of the processes and improve transverse communication regarding engineering, quality and production issues.

Frédéric Henrionnet
CT Infodream

Multi-head processes are a common means of production that enable several parts to be produced in a single process cycle. Examples could include multi-spindle lathes, injection moulding, dosing and foundry applications – basically any process where multiple parts are manufactured at the same time.

Multi-head processes present a challenge in terms of quality assurance and particular care must be taken when devising a method for measuring process capability and ongoing process control using SPC.

The issue is that a multi-head machine needs to be thought of as a ‘single process’ and also as a number of mini elementary processes. Parts from each individual head will contain variation (within head variation) and also parts from one machine head will be different from another (between head variation)

Expanding on this point further, there are process factors (variables) which are specific to each head but also other factors which act on the process as a whole. For example,

Individual head – machine tool, tool wear and tool setting

Whole process – environmental factors, operators, process settings,

As with any other capability analysis and process control system, it is important that we identify and understand these variables and ensure we take them into account when developing our inspection and sampling plan.

Data collection and analysis

As mentioned above, it is important to ensure that when we collect data, our samples contain variation from all the process variables. Sometimes, for example during a process capability study, it may be necessary not to include ‘long term’ variables but from a process control point, the inspection plan should be devised to include all variables.

Once we have collected data, we have 2 options:

• To analyse each head individually
• Global analysis of all heads

To allow us to investigate further we need to ask a key question:The diagram above roughly represents how our data distribution would look for each head and for the global process. As described above, the distribution contains the ‘within head’ variation and ‘between head’ variation.

Two questions now arise:

• « am I able to attach a part to an elementary process? »
• « What type of sampling should I or can I do? »

Am I able to associate a produced part to a specific head?

If the answer is ‘no’, then we must refer to our earlier investigations where we analysed how independent each head actually is from the others, and how many of the process factors are global and how many specific to each head. This will allow us to decide to what degree we can consider the process to be a single machine or a group of sub-machines.

If all the process variables are common and each head cannot be changed independently, then we could consider the machine as a whole and analyse the data for the ‘global process’ without risk of ‘missing’ data.

(In this case, we should still conduct a machine capability study to validate this and ensure parts are included from all heads)

However, if each head is independently variable, then we need to consider the machine as multiple individual processes and each head needs to be analysed and controlled separately.

(That is not to say we cannot take random samples in the future when the capability is known and the process is under control)

Inspection planning and sampling.

The first key point is to ensure that samples are manufactured under conditions which represent all the process variables (as described above).  Assuming this is the case we could inspect parts from a multi-head machine in 2 ways:

• Collection of random parts from the total production
• Systematic sampling of parts from each head

In the first case (random sampling of total production), it is possible to measure the overall process capability but this is not a good method for detecting special causes and controlling the process. This is because grouping samples from each head results in a high variability of the mean (see graph 1) and also potentially a non-normal distribution (because it is basically many normal distributions from each head mixed) (see graph 1).

The high variability will result in control limits which are wider than the ‘true’ control limits of each machine head and therefore, the global control chart will not be very effective at detecting small process changes.

However, if you have decided from earlier analysis that the machine can be thought of as a single processes rather than individual processes,  then it is acceptable to sample global production in this way because the ‘between head’ variation is very small and any changes in the mean value will represent overall process drift (which is want the control chart to monitor)

Also, taking random samples in this way, does reflect the variability which the customer sees in the product and is a true reflection of the overall process capability.

In the second case (systematic sampling of each head), we find ourselves in a more favorable situation and if possible we should use this method even if we consider the machine as a single process. The average distributions of the elementary process should follow a normal distribution, and control charts will be  very effective at detecting special causes as control limits will reflect the historical variation particular to the individual head rather than from the machine as a whole.

As the process capability increases and the process is ‘in control’, there could also be a third case where parts are randomly sampled along with the head traceability or parts systematically sampled from specific heads e.g.for a 64 headed process, we  systematically collect data from heads 1, 12, 24, 32, 48 and 64.  This is basically process sampling where it is possible to get a good reflection of the total process without measuring every head.  (note: this is only possible after all the individual head process factors are being suitably controlled)

Some SPC Software (such as our Qualaxy SPC software) are flexible enough to cater for all these scenarios to provide powerful analysis and control charting to help you understand and control your multi-head machine.

Unitary controlfor each head(8machine heads = 8 control charts)

Global analysis for all heads

Additionally, you can display 2 other types of charts:

Multiple charts  synchronised by time and displaying each head on the same charts

 Transverse chart synchronised by machine head on x-axis

To summarise, we can say that SPC and process capability analysis is a great tool for measuring and controlling a multi-head process.

However, due to the increased complexity we must thoroughly investigate the process variables in order to decide the most effective sampling plan and analysis methods.

The key points are:

• Firstly, do we consider the multi-head machine as a single process (global) or a number of individual sub-processes. To determine this, we need to identify and analyse the process factors which will contribute to both ‘within head’ variation and ‘between head’ variation.
• Secondly, are we able to trace a manufactured part to a specific machine head. Depending on the answer to point one above, this could be critical. If we consider the machine to be multiple processes and we cannot trace the machine head then we will not be able to develop a robust control system.
• Thirdly, how should we analyse the data – globally for the whole machine or individually. Both are important because although the customer sees the overall variation from the process, it is important to control each head to ensure good quality using control charts.
• Finally, SPC software allows you to maximise the charting and analysis possibilities helping you understand and control the process more effectively.

Frédéric Henrionnet
Ct Infodream

References:
– Apply statistical process control SPC / SPC , Maurice Pillet
– The screenshots were done using our Qualaxy SPC software

Attributive control of aspect defects in an SPC approach is done using the classical attribute tracking charts NP (number of non-conforming parts), C (number of non-conformities), P (proportion of non-conforming parts) or U (proportion of non-conformities per unit).

If a C-card is used for high value-added parts in 100% control, the evolution of the number of defects per part will be followed. In some cases, this monitoring is sufficient. On the other hand, one can imagine other cases for which the information on the number of defects is not sufficient.

To overcome this lack, the notion of defect location (or defect map) can be introduced for attributive controls. Instead of only documenting the presence of the defect, the location of the defect on the part is also documented.

The image above shows a well-known possible application of attribute tracking using defect mapping. In the image, a car rental company reports the defects encountered when a customer returns his rental car. In this example, the car hire company finds two defects: a first defect on the bonnet, a second defect on the left front wing.

In the industrial field, one can imagine a multitude of applications that require knowing where a defect is, in order to be able to decide whether or not to reject a part.

In aeronautics, we may accept a crack that is located in an area of low physical stress. On the other hand, the same crack will lead to the part being scrapped if it is located in an area that is subject to high physical stress.

For a watch manufacturer, a scratch can be accepted if it is in a non-visible area of the final part. However, if the scratch is in the middle of the bezel, which is clearly visible to the customer, the same defect would lead to the part being scrapped.

The examples given above show cases where the location of a defect plays a determining role in the decision to scrap or not a part. One can also imagine cases where additional information for the defect (defect length, defect area) is important for the decision.

As with its location, one could ask to specify each defect with a defined amount of additional information.

Depending on the size of the scratch or impact on the bonnet, the lessor’s decision will not be the same: he may accept the return as it is, or he may request a repair and invoke the insurance.

It is therefore up to him to characterise these defects, either by classifying them (small, medium, large), or by measuring them (15mm long scratch).

In some areas, a crack can be considered acceptable if it is shorter than four millimetres. If it is longer, the part should be discarded. For a glass manufacturer, an iridescence can be considered as not disturbing, if its surface does not exceed 5mm².

All these examples illustrate the possibility of using a Defect Map to make the right decisions when it comes to deciding whether or not to scrap a part. If you limit the use of a defect map to this decision support, unfortunately you will not see a reduction in the number of defects.

How can we take advantage of this additional information (location, characterisation) to help reduce the number of defects?

Let’s imagine that our car hire company now « superimposes » the defects found for all the cars hired during a certain period. The result is a defect map that shows in which areas of a car the greatest number of defects are found.

Looking at the above map, he could take corrective action to limit the number of impacts in certain areas and ask for the installation of reversing radars to limit the problems on the rear of vehicles!

Taking this example to the industrial domain, we can see that with the help of the Defect Map, we can easily identify areas that often have defects and look for the cause for this accumulation. By eliminating the identified cause, a reduction in defects is automatically achieved.

By superimposing a production batch (e.g. the 20 pieces of the week) we realise that we have a zone which concentrates a large number of defects (R2-S2 to R3-S5): we would not have realised this by examining the pieces 1 by 1!

To conclude, we can reinforce the statistical aspect of the attribute maps:

• we add information on defects (which allows us to obtain analysis information more quickly, with a smaller number of parts)
• by « superimposing » the Defect Map on the parts, we recreate a sample effect, and we can allow for monitoring and evolution over time.

Frédéric Henrionnet
Quality Assurance Manager, Infodream

Andreas Völker
Product Manager, Infodream

Learn more about Qual@xy SPC
Learn more about Defect Map