The martial arts rank, Shodan, for a first-degree black belt, does not mean “expert”; it means “first step.” ISO 9001:2015 is similarly a valuable and vital first step toward world-class performance, but it is only that—a first step. It covers only by implication many of the risks and opportunities that IATF 16949 covers explicitly, such as six of the Toyota Production System’s (TPS) Seven Wastes as well as crippling supply-chain interruptions, inadequate metrology systems, and more.
IATF 16949 is actually ISO 9001:2015 plus additional requirements, which makes it easy for ISO 9001 users to implement relevant clauses of IATF 16949. ISO 9001 users don’t need to implement all the additional clauses—if they did, they might as well register to IATF 16949—but many of IATF 16949’s key clauses are as relevant to nonautomotive applications as they are to automotive ones. This article will cover what look like the most important ones, although others also might be helpful.
While ISO 9001:2015’s new clause on actions to address risks and opportunities applies to all seven TPS wastes, the standard’s focus is still on only one: poor quality. Poor quality is the easiest waste to detect because it stands up and announces its presence through rework, scrap, or even worse, customer complaints. If the quality is good, the organization often assumes that everything is fine and goes back to business as usual.
Suppose, for example, that masons are laying 125 bricks per hour, which is pretty much the industry’s standard rate, and the walls meet the customer’s requirements. In this case, what we don’t know can and will hurt us. Motion-study expert Frank Gilbreth noticed that the masons were bending over to pick up each brick, to which mechanical engineer Frederick Winslow Taylor added that workers were lowering and raising their entire upper body weight to get a 5-pound brick. The employer couldn’t really pay people to do the equivalent of 125 toe touches per hour, so skilled masons earned paltry wages while the customers paid more than they should have for the buildings. Gilbreth’s introduction of a nonstooping scaffold to deliver the bricks at waist level increased the rate to 350 per hour, and with less overall physical effort. It is doubtful that any losses to poor quality even approached the asymptomatic waste of throwing away almost two-thirds of the workers’ labor.
IATF 16949’s clause 188.8.131.52, which ISO 9001 users won’t find in that standard, cites the need to address the effectiveness and efficiency of each process. These become inputs to the management review and is covered later in clause 184.108.40.206. An effective process delivers the desired results, and an efficient one expends the fewest resources possible to achieve this. As Henry Ford wrote about efficiency almost 100 years ago, “If a device would save in time just 10 percent or increase results 10 percent, then its absence is always a 10-percent tax.”1 ISO 9001’s clause 6.1 on actions to address risks and opportunities relates to this issue only by implication, while IATF 16949 points it out explicitly.
IATF 16949 clause 5.3.2 adds that workers should be empowered to stop the line if necessary to avoid creating poor quality. This practice apparently originated in the automotive industry because workers at Ford’s River Rouge Plant (if not also the older Highland Park plant) had this level of authority 90 years ago, and it was subsequently made famous by Toyota. This is imperative in the rapid assembly-line nature of the automotive industry where, if workers are afraid to stop the line for poor quality, hundreds or even thousands of bad parts can be made in a few hours and then incorporated into the end user’s product. The automotive industry’s approach is to intervene the moment the process makes even one nonconforming item, or even better, the moment an error-proofing (poka-yoke) or autonomation (jidoka) device detects that a nonconforming part is about to be made.
Clause 6.1 is perhaps the best-known improvement in ISO 9001, but it touches only by implication a wide spectrum of risks and opportunities that IATF 16949 identifies explicitly. Most ISO 9001 users recognize that preventive action, which was formerly part of the standard’s “corrective and preventive action,” is now embodied in clause 6.1. IATF 16949 clause 220.127.116.11 requires a process for preventive action, and the Automotive Industry Action Group (AIAG) offers a world-class process: CQI-20, Effective Problem Solving. It is very similar to the Eight Disciplines (8D) process that also originated in the automotive industry and, while designed for corrective action for nonconformances, works perfectly well for preventive action. The only difference between corrective and preventive action is, in fact, the need for a containment step when poor quality is present. Clause 18.104.22.168 also brings in the issue of read across/replicate process, or deployment of lessons learned to similar processes, and cites the importance of organizational knowledge (ISO 9001:2015 clause 7.1.6) in this context.
Next comes the issue of supply chain risks. Supply chain problems terminated Napoleon Bonaparte’s invasion of Russia and similarly mauled an invading Swedish army more than 100 years earlier; it was relatively easy for Tsar Peter I to finish off Charles XII’s already weakened army at Poltava. The Greek mercenary Memnon of Rhodes (portrayed by Peter Cushing in the movie starring Richard Burton) proposed similarly to destroy or at least stalemate Alexander the Great through a scorched earth strategy; Alexander’s elite Companions would have starved to death as easily as any rabble had they insisted on pressing forward. Memnon’s Persian employers insisted on giving battle instead, which is why almost nobody has ever heard of Memnon of Rhodes, the Macedonians used Persepolis as a parade ground, and dozens of cities between Egypt and Central Asia are named Alexandria. Kandahar in Afghanistan is one such city.
The takeaway is, of course, that even if your organization is a corporate Alexander or Napoleon with Six Sigma or even better product realization processes, it can’t make even one unit of product without the necessary inputs—and the automotive industry has known this for decades. Automotive assembly plants that operate on a just-in-time basis can run out of work within hours if deliveries do not arrive on schedule, and Ford had extensive contingency plans to reroute rail shipments if, for example, a flood washed out a bridge. In September 2011 Chrysler lost production at three plants due to a shortage of automobile carpeting when the Susquehanna River flooded a supplier in Bloomsburg, Pennsylvania.2 Pennsylvania had to improve flood protection in the affected area to ensure that this would not happen again.3 Although ISO 9001:2015 does not address supply chain risks explicitly, IATF 16949 clause 22.214.171.124 requires contingency plans to ensure continuity of operations if something happens to the supply chain, or to the utilities and infrastructure upon which the organization relies.
IATF 16949 again goes beyond a focus on quality by requiring, in clause 126.96.36.199 on plant, facility, and equipment planning, the need to optimize material flow and handling, and also to apply lean manufacturing principles. Inventory, one of the TPS’s Seven Wastes, is proportional to cycle time, and poor layouts that result in extensive material transportation result in nonvalue-adding cycle time.
This clause of ISO 9001:2015 requires that gauges and instruments be calibrated (or verified) to ensure their suitability. There is, however, no explicit provision for gauge precision, or the gauge’s ability to get the same measurement from the same part consistently. An inadequate gauge or instrument, at least in terms of precision, can hide in plain view in a fully functional ISO 9001 quality management system. IATF 16949 clause 188.8.131.52.1 requires measurement systems analysis (MSA), also known as a gauge repeatability and reproducibility (R&R) study.
MSA quantifies the inherent gauge variation (equipment variation or repeatability) and also appraiser-dependent variation, or reproducibility. Many ISO 9001 users rely on off-the-shelf calibration management software such as GAGEtrak and GAGEpack to ensure that gauges never miss their calibration schedules. GAGEtrak also includes “conduct gage R&R analysis (MSA fourth edition),” while GAGEpack “…provides all of the tools you need to create a complete statistical and graphical analysis of your measurement system.” In other words, if you’re using one of these packages (and there are probably others) to manage your gauge calibration schedule, you already have a system in place to handle your R&R studies as well. Statgraphics and Minitab also handle the calculations for R&R studies, although they are not designed for calibration schedule management.
MSA is one of AIAG’s core tools, and Measurement System Analysis, fourth edition, is the AIAG’s most recent manual for gauge studies. It includes not only the procedure for a good gauge study but also a list of factors that can affect gauge performance, and procedures for gauge studies that involve destructive testing and attribute testing.
Ford wrote that many, if not most, of his company’s amazing productivity improvements were initiated by frontline workers rather than management. The idea that the people who have their hands on the job eight (or more) hours a day know more about it than anybody else is at least a hundred years old.
“In one shop in the Middle West where there is a large measure of self-government, the employees voted to retain an expert, and later they posted a sign announcing that they were all ‘efficiency engineers’ themselves! Acting on that principle, they have each studied their job in light of the best practice; and as a result that factory, with an actual decrease in the number of men working and without additions to equipment, has more than doubled its output.”4 IATF 16949 deploys this principle through clause 7.3.2, which requires a documented process for employee motivation and empowerment.
The fact that ISO 9001 no longer requires a quality manual does not mean it is not a good idea to have one. It is a convenient place to organize the entire quality management system through a list of processes along with matrices of their interactions and handoffs, the scope of the QMS, the context of the organization, needs and expectations of relevant interested parties, and the quality policy. IATF 16949 clause 184.108.40.206 requires a quality manual, although it can be a series of documents rather than a single document.
IATF 16949 clause 220.127.116.11 adds advanced product quality planning (APQP), which is known outside the automotive industry as advanced quality planning (AQP). APQP is another AIAG core tool for which AIAG offers a manual; it also includes guidance for control plans. D. H. Stamatis’ Advanced Quality Planning (CRC Press, 2018) also is an excellent reference that discusses quality planning and control plans.
IATF 16949 adds design for manufacture, a principle that also dates back to Ford, and failure mode and effects analysis (FMEA) which is yet another AIAG core tool. AIAG has added considerably more sophistication to the FMEA process with the fourth edition of its FMEA Handbook and most recently the AIAG & VDA FMEA Handbook. IATF 16949 clause 18.104.22.168(j) also raises the issue of standard work, which is itself an important tool not only to ensure consistent results but also to promote continual improvement.
IATF 16949 clause 22.214.171.124 adds supplier quality management system development, which includes encouraging suppliers to register to ISO 9001 and hopefully IATF 16949. A strong argument could be made for also encouraging suppliers to register to ISO 14001 and ISO 50001 to reduce material and energy wastes, respectively, as well as associated supply-chain costs. Clause 126.96.36.199 adds supplier development, which is again nothing new. The C. R. Wilson Body Co. wanted the Ford Motor Co. to pay $152 per car body, or roughly 30 days’ worth of wages as paid by Ford. Ford said he would pay $72, and Wilson protested that it cost more than that to make them. Ford’s production chief Charles Sorensen showed Wilson how to get its costs down to $50 per body while paying higher wages.
As Ford wrote of supplier development, “Under the pressure of necessity, he found he could make cost reductions here, there, and everywhere, and the upshot of it was that he made more money out of the low price than he had ever made out of the high price, and his workmen have received a higher wage.”5 The idea is not to squeeze suppliers for lower prices but rather to show them how to remove waste (muda) that adds no value for anybody while driving prices up, and wages and profits down.
IATF 16949 clause 188.8.131.52 adds explicitly the need for process failure mode effects analysis (PFMEA), which identifies activities for which control plans must be defined. Appendix A of the standard provides additional guidance on control plans. Clause 184.108.40.206 relates again to standardized work, 220.127.116.11 adds total productive maintenance (TPM), and 18.104.22.168 addresses production scheduling and just-in-time manufacturing. These considerations go well beyond quality assurance to support continuity of operations (through avoidance of unplanned shutdowns) and minimization of inventory, which is another of the TPS’s Seven Wastes. IATF 16949 also expands considerably on ISO 9001:2015 clause 8.5.2, Identification and traceability.
Any change to a process, and this can include a startup after a shutdown, or a changeover, introduces the chance of unexpected and undesirable consequences. The chemical-process safety principle is known as management of change (MOC), although it applies equally to quality. IATF clause 22.214.171.124 requires a documented process to address this issue. What happens when we change tooling, materials, suppliers, personnel, or methods? What happens when a process starts up after a shutdown for maintenance or other reasons?
IATF 16949 clause 8.6.6 requires an acceptance level of zero defects for attribute data. As far as I know, this cannot be achieved by any sampling plan, such as those defined by ANSI/ASQ Z1.4, for which the lowest AQL is 0.01 percent. A sample is, almost by definition, less than 100-percent inspection. Clause 8.6.6 cites, however, clause 126.96.36.199, which discusses process capability studies. A Six Sigma process that is centered on its nominal will deliver two parts per billion nonconforming if the process is normally distributed. Shigeo Shingo’s Zero Quality Control: Source Inspection and the Poka-Yoke System (Routledge, 1986) adds successive check systems, error-proofing systems, and self-check systems. Error-proofing or poka-yoke prevents the generation of nonconforming work, while self-check systems intercept it before it can leave the workstation. The key takeaway is, however, that nonconformance levels on the order of whole percentages or even fractional percentages that might have been taken for granted a few decades ago are no longer acceptable today.
IATF 16949 clause 188.8.131.52 adds as inputs for management review the cost of poor quality (for which AIAG’s CQI-22 The Cost of Poor Quality Guide is a good reference) and also process effectiveness and process efficiency. This makes it clear that management review must address wastes other than those of poor quality.
Deficient corrective and preventive action (CAPA) is one of the biggest sources of ISO 9001 audit findings6 along with FDA Form 483 citations.7 Inadequate CAPA and inadequate problem-solving processes are the two primary sources of major IATF 16949 nonconformances and comprise roughly one-third of them.8 ISO 9001:2015 does not contain an explicit requirement for a problem solving process. IATF 16949 clause 10.2.3 does, and AIAG’s CQI-20 Effective Problem Solving is an outstanding off-the-shelf resource for it. Chris Visser’s 8D Problem Solving Explained(CreateSpace Independent Publishing Platform, 2017) is an excellent reference for the very similar 8D process.
CAPA, of which problem solving is a central element, is arguably the most important process in any quality management system. It addresses, or at least it ought to address, issues far beyond its traditional role of prevention or correction of poor quality, the most obvious of the TPS’s Seven Wastes. The same process can address customer complaints, audit nonconformances, risks and opportunities identified by the management review meeting, and the other six TPS wastes.
The basic idea is to treat any gap between the current state and an ideal or theoretical future state as a “nonconformance” on which the CAPA process can operate, although in the absence of poor quality, containment is not required. Suppose, for example, that 100 lb of stock go into a machining operation, and 40 lb of product come out. The fact that 60 percent of the stock is wasted (although it might be recyclable) is a gap between where we are now and where we could possibly be. We can use this as a problem definition on which our CAPA process can work, even if the process is not generating any nonconforming work to contain.
CQI-20 adds the need to address not one but three root causes of poor quality:
1.The escape root cause is the reason the poor quality reached an internal or external customer. If the poor quality was intercepted by the process controls (such as those defined by the control plan), there is no escape root cause.
2.The occurrence root cause is why the poor quality was produced in the first place. This is the traditional focus of CAPA.
3. The systemic root cause is why the quality system or planning process did not identify the issue ahead of time. As but one example, failure mode and effects analysis is intended to anticipate trouble and identify suitable controls to at least detect it and, ideally, prevent it. IATF 16949 clause 10.2.3(d) adds explicitly the need to deploy the results of a CAPA project to similar activities and processes that might benefit from the knowledge. Appendix B of CQI-20 adds that failure to do this for similar products and processes is an inhibitor to effective problem solving, which underscores the importance of not only the organizational knowledge called for by ISO 9001:2015 clause 7.1.6, but also a step in the CAPA process to ensure that this happens. Ford put the idea much more simply: “An operation in our plant at Barcelona has to be carried through exactly as in Detroit—the benefit of our experience cannot be thrown away.”9
IATF 16949 clause 10.2.4 requires a documented process for error-proofing or poka-yoke, which is an appropriate process control for incorporating into a control plan. Useful references include AIAG’s CQI-18, Guideline to Effective Error-Proofing, and Shingo’s Zero Quality Control: Source Inspection and the Poka-Yoke System. Productivity Press’ Mistake-Proofing for Operators: the ZQC System is based on Shingo’s book and is easy for frontline workers to understand.
IATF 16949 clause 10.3.1 requires a documented process for continual improvement. This process should focus on reduction of waste (muda) and process variation. Note again that IATF 16949 looks at all forms of waste rather than just poor quality. A strong case can be made for reduction of variation in processing and transportation times as well as product characteristics. Eliyahu Goldratt’s and Jeff Cox’s The Goal (North River Press, 30th anniversary edition, 2014) showed how variation related to production control increases cycle time and therefore inventory, even if there is nominally enough capacity to handle the work.
The Goal’s matchstick-and-dice simulation showed why it is “impossible” to operate a balanced factory at close to 100-percent capacity. Every operation is a capacity-constraining resource under these conditions, which means variation will alternately starve and overload each constraint; this leads to bubbles of inventory. (If you have bubbles of inventory moving around your factory, this is probably why.)
In his 1922 book My Life and Work, Ford claimed, however, to have achieved what Goldratt showed much later to be “impossible”:
“The idea is that a man must not be hurried in his work—he must have every second necessary but not a single unnecessary second.” This suggests that every operation at Ford did operate at close to 100-percent capacity, and apparently without generating inventory bubbles. The only way Ford could have achieved this was to eliminate almost all variation in processing and material transfer times. He designed his production system to act much like a clock, with synchronization of production to deliver to each product realization process what it needed and when it was needed, and neither too early nor too late. Automation removes human-induced variation from the job. Standard work, which defines, among other things, the takt time for production, also suppresses the variation that Goldratt described.
Kanban and other just-in-time systems also are useful, while it can be argued that Goldratt’s drum-buffer-rope (DBR) production management system contains but does not remove the effects of variation. DBR does, however, prevent the capacity-constraining resource from running out of work, and also avoids accumulation of inventory anywhere but the buffer that protects the constraint from running out of work.
Ford’s four key performance indicators (KPIs) are meanwhile a good starting point for continual improvement, since we cannot improve anything until we recognize the opportunity. Ford stated that it is possible to waste time, material, and energy. Time can be broken down into time of things (cycle time) and time of people (waste motion), for a total of four KPIs that everybody can easily understand. All seven TPS wastes can be expressed in one or more of these terms. Inventory is, for example, proportional to cycle time, while wasted motion wastes the time of people. Ford wrote of the latter that nobody can be paid to walk, and no job should require anybody to take more than one step in any direction or bend over.
Peter Gaa’s Creating and Using a Shingo-Style Process Map (Rikai Publishing, 2018) shows how to create process maps that are 100-percent consistent with the process approach of ISO 9001 and IATF 16949, and these process maps force nonvalue-adding cycle time to become visible. We should be able to look at any process and determine (e.g., from standard work, which IATF 16949 cites explicitly) how much time the process actually requires to transform the product. If, for example, a machine tool is in contact with the part for five seconds, but the work spends five minutes in the operation, we know immediately that most of the time is consumed by 1) setup and handling; 2) waiting; 3) inspection; and/or 4) transportation. The work cells that originated at Ford Motor Co. reduced transportation time to almost nothing, while unitary machines reduce setup and handling. “Wait” is indeed a four-letter word, even though it is legal to say it on the radio because it increases cycle time and inventory while adding no value whatsoever.
If one draws a control surface (i.e., analytical boundary) around any process, it is possible to perform a material and energy balance from which no material or energy waste is likely to hide. If, for example, 100 lb of a consumable like paint (including the solvent) goes into a spray booth, and 2 lb of paint solids come out on the product, we ask immediately what became of the other 98 lb of solids and solvent. Shingo discovered a situation of this nature in which overspray was not only wasting paint, it was contaminating groundwater and causing a problem for nearby rice farmers. He then asked the plant manager whether the objective of the process was to paint the parts or the air, and corrective action was taken for a problem that had previously hidden in plain view. Ford recovered solvents from coating operations so efficiently that one gallon did the work of 10 before it finally dispersed into the atmosphere. The material and energy balance therefore not only forces these wastes to become visible, whereupon the organization’s CAPA process should deal with them in short order, it also supports ISO 14001 and ISO 50001.
-WILLIAM A. LEVINSON