9. Improving Changeover
9.1 Some Common Issues Associated with Changeover
People are often concerned about the length of time required for changeovers while failing to get a proper grasp of the situation. The task is left up to the operators, with problems concerning the items listed below left unsolved. The result is unstable changeover times with no understanding of the reasons behind their wide variation.
- Working methods (procedures, methods, operator skills)
- Jigs and tools (shape, mechanism, precision)
- Equipment precision (precision requirements, relationship between precision and adjustment)
- Engineering issues (technical improvements required)
- Supervision (need for evaluation)(2) Unclear proceduresAlthough the sequence of operations is the most difficult thing about a changeover, in many cases there is no set procedure, and each operator does the changeover his or her own way, using different methods, a different sequence of steps, and different adjustments. This is why changeover times vary so widely and why, depending on the operator, problems may be experienced when production is started up, making it necessary to reset the machines. The procedures may be unclear either because none have been specified or because they are not properly taught.(3) Lack of investigation of adjustment, and failure to achieve right-first-time changeoverAlthough adjustments account for around 50% of changeover time, the problems associated with them tend not to be investigated thoroughly enough (see Table 4.14). Adjustment is regarded as a difficult subject and a necessary evil, and few attempts are made to improve it. However, although not all adjustments can be eliminated, some certainly can, so the first requirement is to study the mechanisms governing them and work out which are avoidable and which are not. The avoidable ones should then be eliminated, and the unavoidable ones speeded up. Reducing the time taken for adjustments is a crucial part of changeover reduction, since they account for such a high proportion of the total time.
Table 4.14 Percentage of Total Time Taken by Typical Changeover Tasks
Although it is theoretically possible to achieve right-first-time changeover (changeover that requires no trial processing) by eliminating adjustments, this has rarely been achieved in practice. Nevertheless, right-first-time changeover is possible even in the machining and assembly industry, provided that the product is accurate to no less than 5 μm. Achieving it would be highly beneficial, so it is a major issue that should be tackled without delay.
9.2 Strategies for Improving Changeover
(1) Differentiate between internal and external changeover
External changeover tasks are those that can take place while the equipment is still in operation. They include preparations such as getting jigs and tools ready, preparing storage areas, trolleys, etc. for parts to be taken off the machine, pre-assembly, and preheating. Internal tasks, in contrast, can only be performed with the equipment shut down (e.g. replacing parts, centering, and adjusting). When improving a changeover, it is essential to work out which of the tasks can be done externally and what has to be done internally, and establish standardised procedures for carrying them out. (N.B. the terms ‘internal’ and ‘external’ were coined by Shigeo Shingo, a Japanese manufacturing expert who developed the concept of SMED, or ‘single-minute exchange of die’).
Someone watching a changeover will usually see many delays that the person performing the changeover does not perceive as problems – looking for a missing tool, for example, or going to fetch a bolt because the one at hand does not fit. Although the time wasted on each occasion might be as short as one or two minutes, it is essential to eliminate these delays, because the total time lost can be very great.
Changeover time can be substantially reduced by asking the following questions in advance and making the necessary preparations:
1 What preparations can be made before starting the changeover?
- What jigs and tools are needed?
- What parts are necessary? How many are needed?
- Are the parts to be installed in good repair?
- What type of workbench or work area is needed?
- Where should parts be placed after they have been removed?
The following three simple rules should be kept in mind in improving changeover:
- Don’t waste time looking for anything (materials, parts, tools, etc.).
- Don’t move around unnecessarily (position workbenches and storage areas so as to minimise movement).
- Don’t use the wrong tools or parts (using whatever tool happens to be at hand is a common cause of having to repeat the task).Ensuring that these rules are followed requires faithful application of the industrial housekeeping principles known as the 5 Ss. These are seiri (sort), seiton (store), seiso (shine) seiketsu (standardise) and shitsuke (sustain). ‘Sort’ means sorting out what is needed from what is not, and disposing of anything unnecessary so as to free the area of clutter; ‘store’ means working out the most efficient way of storing or positioning the tools and materials needed; ‘shine’ means keeping them clean and ready for immediate use; ‘standardise’ means establishing rules for ensuring that the first three Ss happen; and ‘sustain’ means rigorously following these rules. Following these principles before, during and after changeovers will ensure that the necessary parts and tools are always stored in the designated place in the required numbers and that they are always ready for use, so that no time is wasted while the machine is not making product.
2 Separating internal and external setup tasks
The first step in improving a changeover is to examine the overall procedure as currently performed and answer the following questions:
- Which tasks should be external, and what sequence should they be done in?
- Which tasks should be internal, and what sequence should they be done in?Each individual task should then be scrutinised, and the following questions answered:
- Is it really necessary?
- If it is currently necessary, is there any way of eliminating it?
- Is there any better way of doing it than the current one?
- Can it be simplified?
- Can it be integrated with another task?
- Can it be made more consistent?
- What improvements need to be made?
- What are the key points for performing it effectively?The sequence in which the tasks are performed should then be examined, and the following questions answered:
Can the current sequence be improved?
Can any of the steps be combined?
Should the steps be performed in a different order? Can any of the steps be performed in parallel?
Finally, the way in which the tasks are apportioned should be re-examined, and the following questions answered:
- Are the tasks divided up in the best way?
- Is the changeover being performed by the right number of people?
Implementing the improvements suggested by this line of questioning can reduce changeover time by 30% to 50%. Although separating external and internal tasks is a fairly elementary approach, it is surprisingly effective when combined with establishing clear procedures for individual tasks. If these two actions are done properly, the variability in changeover times can be considerably reduced.
The aims of improving changeovers in this way are to:
- Eliminate problems occurring after startup.
- Standardize changeover procedures so that anyone can complete them within a
- Standardize changeover procedures so that anyone can complete them within a
- Identify any mechanical problems with jigs, tools and methods.
Like other improvements, changeover reduction must be implemented step by step. Making improvements at random, without following the procedure explained above, will not be very effective.
(2) Convert internal tasks to external ones
Examining the tasks involved in a changeover and working out how to convert internal ones (ones performed with the equipment shut down) into external ones (ones performed while it is running) is a powerful method of reducing the changeover time.
For example, a jig that is usually changed, assembled, and adjusted while the machine is stopped can be pre-assembled while the machine is still operating, or parts that are usually adjusted on the machine can be adjusted off-line using standard gauges. Methods such as the following can be used to convert internal tasks to external ones:
1 Pre-assemble. Rather than installing several parts one by one while the machine is stopped, assemble them beforehand and replace the entire assembly.
2 Use standard, quick-fitting jigs. Compare the shapes of the jigs used for different products and see whether any of them (or any part of them) can be standardised for use across the product range, and whether quick-release mechanisms can be incorporated that enable them to be replaced with a single action.
3 Eliminate adjustments. Wherever possible, make adjustment an external task.
4 Use intermediary jigs. Whenever a cutting tool is changed, it must be centered.
However, the need for centering can be eliminated by setting the tool in a standard, high-precision intermediary jig which is installed on the tool post instead of mounting the tool in the toolpost directly.
(3) Shorten internal tasks
Internal tasks themselves can often be shortened by using standard, quick-fitting jigs and better assembly and fastening methods, and by eliminating adjustments.
1 Improve clamping mechanisms. Reduce the number of bolts, for example, or replace screw-type clamping mechanisms with hydraulic ones.
2 Work in parallel. Two people working simultaneously can sometimes complete a changeover more efficiently than a single person performing one step at a time. Although timing and coordination are crucial to the success of this approach, changeover times can often be halved with the same number of labour-hours, even in the most difficult cases.
3 Use the most appropriate number of workers, and allocate the tasks in the best way. Complex changeovers must sometimes be performed by dozens of people. In such cases, considering the following points can reduce setup and adjustment time drastically:
(a) What is the optimal number of workers for each task?
(b) How should the tasks be shared?
(c) What is the critical path for the changeover? What are the constraints preventing it from being shortened? (use of crane? labour resources?)
(d) What is the best way of using the labour available?
Table 4.15 shows some examples of strategies for improving setup.
(4) Eliminate adjustments
Many adjustments can be converted into ‘condition-setting’; that is, setting the required conditions in one go, without the need for trial and error. First, all the current adjustments should be examined to determine whether they are avoidable or not by looking at what their purpose is, what gives rise to the need for them, what exactly they consist of, and how effective they are. The items that should be taken into consideration when doing this are explained below.
Adjustments are usually performed for one or more of the following purposes:
- Positioning: adjusting height, or position on the X or Y axes (for example, setting a press die on a bolster and adjusting its height).
- Centering: centering cutting tools, workpieces, etc.
- Dimensioning: adjusting the depth of cut of a machine tool to achieve the specified dimensions, for example.
- Timing: adjusting the timing of various equipment mechanisms.
- Balancing: adjusting lateral pressure, balancing with set screws or springs, etc.
Adjustments are needed in the following circumstances:
(a) Accumulated errors
This is the most common reason why adjustments are needed, when all the small individual errors in the equipment or the jigs and tools add up to produce a large error necessitating adjustment. Sometimes the error is just in the machine itself, but more usually it is a combination of the cumulative errors in the machine and the associated jigs and tools.
(b) Lack of standardisation
Adjustment is required when there is insufficient standardisation- for example, when reference surfaces are not set, dimensions are unquantified, the amount to be removed in one pass of a grindstone is not specified, etc.
(c) Insufficient rigidity.
Even if everything is satisfactory when the machine is stationary, errors may be produced if the equipment flexes during operation because it is not stiff enough.
(d) Mechanical deficiency.
Some mechanisms are designed to be adjusted by a human operator and would have to be redesigned to eliminate it.
As stated earlier, right-first-time changeover produces good product from the start, without the need for trial processing. This is radically different from the usual type of changeover, where the parts are changed, some product is produced and checked, adjustments are made based on the results of the checks, more product is produced and checked, further adjustments are made, and the process is repeated, usually at least two or three times, until the product satisfies the specifications. In right-first-time changeover, the parts are changed and all the settings are mechanically locked into position, so that no adjustments are required (see Figure 4.30).
The effectiveness of adjustment tasks should be analysed in order to determine whether they are really necessary or not (see Table 4.16 and Figure 4.31). The procedure described below should be followed in doing this:
Table 4.16 Analysing Adjustment Effectiveness
Consider Possibility of Elimination
Understand the purpose of the adjustment
Find out exactly how the adjustment is performed
Find out why the adjustment is considered necessary at present
Identify the function served by the adjustment
Find out what makes the adjustment necessary
Consider whether the adjustment is avoidable or unavoidable
(a) Identify purpose.
Find out why each adjustment is being performed, and confirm the true purpose if it is unclear. Remember that some adjustments will have more than one purpose.
(b) Check details.
Find out exactly what the current adjustments entail (procedure, methods, standard values, number of repetitions, key points, difference between coarse (initial) and fine (final) adjustments, clamping methods, adjustment mechanisms, reference surfaces, functions of adjustment (whether single or multiple) measurement methods, movement methods, interrelationships between adjustment points (whether independent or interrelated), etc.
(c) Understand reasons.
Use the insights gained in the preceding analysis to find out why each procedure is currently needed. Consider the operations individually and in groups, investigate the aims of each in detail, and list the apparent reasons for the procedures.
(d) Analyse principles.
Consider the principles behind the procedures and identify the real function served by the adjustment. For example, does it match heights, make left and right parallel, level something, position something on the X and Y axes, or perform some other function?
(e) Investigate causes.
Use the results of the previous step to identify why the adjustment is necessary. Is the adjustment necessitated by an accumulation of errors, lack of standardisation, insufficient, rigidity, mechanical deficiency, or something else? The need for adjustment may be due to one or more sets of circumstances.
(f) Consider possibility of elimination.
Finally, work out whether or not each adjustment is avoidable or unavoidable. Adjustments due to an accumulation of errors (in equipment, jigs and tools, or assembly) or to lack of standardisation (reference surfaces unspecified, or distances from reference surfaces not standardised) are often avoidable.
(5) Improve unavoidable adjustments
If an adjustment cannot be eliminated, several strategies are available for making it more efficient:
1 Use numerical settings
Wherever possible, avoid adjustments by using discrete numerical settings. Where this cannot be done, consider what measuring methods could be used to eliminate the need for trial and error, or whether a substitute characteristic could be used for making the adjustment.
2 Establish a standard procedure
Establish a standard procedure for performing the adjustment and make sure all the steps are thoroughly understood. After each step, the person performing the adjustment should ensure that the correct result was produced and that the adjustment was within the correct range before proceeding to the next step. Also, consider how adjusting one part of the machine affects other parts, and look for ways of minimizing an adjustment’s effects on other quality characteristics and making subsequent adjustments easier.
3 Improve skills
It is important to increase workers’ skills by having them practice the procedures. Repeating the tasks until they are embedded in the workers ‘muscle memories’ is the key to retaining them over long periods.
9.3 Changeover improvement procedure
Figure 4.32 gives a systematic overview of the changeover improvement procedure.
Figure 4.32 Changeover Improvement Procedure
10 Reducing Startup Losses
10.1 Some Common Issues Associated with Startup Losses
(1) Weak theoretical approach
The problems that happen at startup – frequent adjustments necessitated by
variability in dimensions, long cycle times on grinders with positioning devices installed, tool breakage because of abnormal machine movements and so on – are usually obvious, but the causes of these problems and the mechanisms by which they occur, the associated thermal displacement curves, and the times required for the operation to stabilise are often not considered. They tend to be ascribed simply to changes in the temperature of hydraulic fluid, lubricant or coolant, but they ought to be investigated much more thoroughly, by seeking answers to the following questions:
- What is the significance of the thermal displacements?
- Where do they occur, which direction do they act in, and how can they be measured?
- How many workpieces have to be produced before the dimensions stabilise, if no adjustments are made?
- What is the minimum idling time required?
- How does the thermal displacement of each section of the machine change over time (what do the thermal displacement curves look like)?
- What are the machine tool manufacturer’s views on the thermal displacement, and to what extent has it been suppressed?
- What actions have the operating teams taken to deal with it, and by how much have they reduced it?Few companies are yet able to answer these questions satisfactorily, because they are not taking a theoretical approach to addressing thermal displacement or are not even measuring it. The first thing that needs to be done to ensure smooth startup in the morning and bring the lost time down as close as possible to zero is to investigate and take measurements on the items listed above. Measurements need to be taken in order to determine what kind of thermal displacement is occurring, how it changes over the period between startup and machine stabilisation, how the temperature of the hydraulic fluid changes, whether the lubricating oil has any effect (such as causing the work table to rise or sink), and so forth. Morning startup losses remain at high levels because the measurements described are difficult to perform, the losses have not been accurately identified, and the pace of improvement is slow. These losses need to be identified and reduced as a matter of urgency.
(2) Lack of standard values
Machines are usually run without product for a certain length of time after startup in order to minimise losses, but it is often unclear exactly how long they need to be idled for. There is no hard data to indicate whether 10 minutes, 30 minutes or some other length of time is required, so it is left to the operator’s discretion. If the idling is necessitated by thermal displacement, then it is to some extent unavoidable, but at least a standard minimum time should be established, and steps taken to ensure that it is complied with.
Startup losses tend to be investigated less thoroughly than the other 6 Big Losses because reducing them requires the application of specialised engineering technology and because people are still generally unaware of their importance. A latent time loss of one hour per week may not seem very much compared with the total operating time, particularly if the other 6 Big Losses are still very large; but it will begin to look more and more significant as the other losses decrease. The first thing that must be done is to make the loss visible.
10.2 Strategies for Reducing Startup Losses
(1) Start with measurements
To reduce startup losses, the thermal displacement must first be measured. Thermal displacement is a phenomenon caused by the thermal changes that take place when equipment is started up or shut down, in which some of the components expand or contract, causing them to elongate or shrink in the X, Y or Z directions. As a result, the processing point (the contact point between tool and workpiece), which is determined by the relative position of the workpiece and the cutting tool or grindstone, is displaced (see Figure 4.33). The displacement of the main shaft, grindstone, spindle, table, etc. should be measured during a continuous production run of 300 products beginning from a cold start (after the equipment has been allowed to cool down naturally for a long time from its warm state just after shutdown, when its temperature was somewhat higher than ambient). Each section of the equipment should also be measured at intervals of 0.5 hours, 1 hour, 2 hours, 5 hours and so on as it cools down from its temperature at shutdown until it is completely cold, in order to determine whether or not any displacement occurs and measure it if it does.
Figure 4.33 Thermal Displacement of Spindle Heads and Change in Outside Diameter
The thermal displacement of each section should also be monitored after the equipment is started from cold on a Monday morning, until all the different parts have warmed up and the quality of the product has stabilised, to see whether it is the reverse of what happens when the equipment is allowed to cool down. The following measurements should be taken:
On the workpiece:
- Change in workpiece dimensions from first piece onwards without adjustment (this will inevitably generate defectives)
- Change in workpiece dimensions from first piece onwards with adjustment
- Mean time between adjustments
- Change in Cp value with time after startup
On the equipment (taking a grinding machine as an example):
- Change in temperature of main spindle
- Displacement of main spindle in X, Y and Z directions
- Displacement of work table in X, Y and Z directions
- Displacement of grindwheel in X, Y and Z directions
- Displacement of grindstone dressing device
- Change in temperature of hydraulic fluid
- Change in temperature of lubricantBefore performing these measurements, careful consideration should be given as to what type of measuring instruments to use, what to take as a reference surface, how accurately to measure, and how to record a series of measurements over time (whether manually or automatically). The results of the measurements should be used to determine the following:
- Change in dimensions of workpiece over time, and time taken to stabilise
- Relationship of the above with displacement of each section of the equipment
- Provisional standard for minimum warming-up time needed under present conditions
- Cause of thermal displacement of each section of the machine
(2) Review materials
According to theoretical principles, thermal displacement is to some extent unavoidable because of the coefficients of thermal expansion of the materials involved. Nevertheless, the materials should be reviewed with the aim of minimising the displacement or the time taken for the machine to stabilise. One approach is to discuss the following points with the machine manufacturer:
- Has the machine been designed with thermal displacement in mind?
- What is the thermal displacement of each section of the machine? Are thermal displacement curves available?
- Have the materials or systems been selected with a view to cancelling out thermal displacement?
- What should be done to ensure that the machine stabilises as quickly as possible?
The following questions concerning structural materials should also be investigated:
- What are the components of the relevant sections of the machine, and what materials are these components made from?
- What are the coefficients of thermal expansion of those components?
- Do the measured thermal displacement curves closely match those expected from materials with these coefficients of thermal expansion undergoing these temperature changes?
- Could any other materials with lower coefficients of thermal expansion be used, and, if so, would they be strong enough?
(3) Cool the relevant parts
As explained, thermal displacement is caused by local temperature differences within the machine, and these temperature rises can be prevented by physically cooling the relevant parts. The following specific actions can be taken:
- Give the machine itself a thermally-radiative structure: expose heat-generating motors and rotating parts (spindles, etc.) and equip them with cooling fins.
- Construct the equipment in such a way that it can be force-cooled by air or water (devices such as boring heads, for example, are often fitted with water-cooled jackets).
- Control lubricant temperature: lubricant can be circulated through heat-generating sections of the machine to extract the heat, and then passed through a cooler.
- Use cutting fluid as a coolant: control the temperature of coolant in the same way as that of lubricant.
(4) Use an automatic compensation system
Given that thermal displacement is to some extent unavoidable, one way of dealing with it is to compensate for it automatically in order to achieve stable production right from the start. It is easy to programme automatic compensation into the software of numerically-controlled tools and machining centres, using a pattern like the following:
- Estimate the displacement in advance
- Set the compensation at a high value of 50 μm for the first 2-3 pieces
- Reduce it to 30 μm for pieces 4-6
- Reduce it to 15 μm for pieces 7-9
- Reduce it to 8 μm for pieces 10-12
- Then produce as normal, without compensation
This approach can be used when the thermal displacement curve is measurable and consistent, and there is complete data for every season.
Table 4.17 summarises the approach to reducing startup losses described in this section.
Table 4.17 Procedure for Reducing Startup Losses
11 Reducing Minor Stops
11.1 Minor Stops and Idling
When a problem occurs, a machine may either stop completely for a short time or run empty until it is reset. Stopping completely for a short time is called a ‘minor stop’, and running empty is called ‘idling’, but, for the purpose of this discussion, both of these will be lumped together under the term ‘minor stops’.
(1) When the equipment stops completely
A minor stop may occur when a sensor detects a problem and automatically stops the equipment. Some examples of such problems are:
- Overloading. Automatic packing and assembly machines often stop because of overloading, which can happen when products collide with each other, for instance.
- Quality problems. Automatic assembly machines, transfer machines and other automatic machines often stop when a sensor trips because of a quality problem – for example, when a part drops off a suction pad and an assembly error occurs.
(2) When the equipment runs empty
Equipment runs empty, or idles, when materials stop coming but the machine itself keeps running. This can happen when the mechanism of the equipment makes problems difficult to detect, or if sensors are too costly to install, and it is often not noticed for quite some time. Idling commonly occurs in automatic machinery because of problems with the mechanisms that feed or transport the work.
11.2 Characteristics of Minor Stops
(1) Because minor stops are so easily dealt with, radical solutions are not sought
Being so easily corrected, minor stops tend not to be regarded as a problem. As a result, no permanent solutions are sought, and they are simply tolerated.
(2) They occur in various different ways
Minor stops may occur with certain products or parts but not others, or only on certain machines on certain days. Their apparently random nature inevitably makes them easy to overlook.
(3) Their location constantly changes
Minor stops seldom occur at the same location on a machine. They are more likely to be concentrated in one area at one time and in a different area at another, making it hard to pin them down. They may also be purely chronic, or a mixture of sporadic and chronic.
Minor stops are purely chronic when they occur in one part of the machine, various steps are taken to solve them, and they disappear from that part of the machine only to reappear in another, leaving the overall situation much the same. This happens when attention becomes focused exclusively on one part of the machine while other parts, despite having hidden problems, are ignored. The key to reducing minor stops is to uncover and eliminate every latent defect in any part of the equipment where minor stops might occur, regardless of how often they actually do.
Sometimes, a sporadic minor stop will occur at the same time as a chronic minor stop. The sporadic stop could be due to a faulty part or an incorrectly installed jig; but whatever its cause, it is a different phenomenon from the chronic one and manifests
itself in a different way. It is important to distinguish between them quickly and take the appropriate corrective action.
(4) They are difficult to quantify
Although logging details of the location of minor stops, the number of times they occur, and the time taken to correct them, etc., is practicable over short periods, it is much more difficult to do so continuously for a long time. It is possible if each operator or operating team looks after a single piece of equipment, but difficult when they are responsible for several machines. Although the losses due to minor stops can be calculated from the net operating time and the output, this does not convey anything about the actual number of occurrences. Some companies install automatic counters to obtain this information.
11.3 Some Common Issues Associated with Minor Stops
(1) The losses remain hidden
Minor stops may be simple and easily corrected, but the losses they generate are surprisingly large if their rate of occurrence is high or if it is a long time before they are noticed. They are often not treated as a serious problem because people are unaware of the big losses they cause; however, in the case of automatic assembly machines and other automatic equipment, they rival changeover in accounting for the largest (20% – 40%) proportion of total losses. The first step in addressing them is therefore to measure the size of the losses.
Although it is important to find out how many minor stops occur each day or shift, either the mean time or the number of machine cycles between occurrences should also be monitored. In a high-speed machine, it is better to monitor the mean number of cycles between occurrences; often this is 100-200 before improvement but greater than 10,000 after.
(2) The causes are not addressed satisfactorily
Minor stops are not usually given the attention they deserve. They are only examined superficially and are typically dealt with by stopgap remedies that do not go the whole way and only address part of the problem, treating the symptoms but not the underlying causes.
(3) The phenomena are not observed closely enough
The most important thing to do when trying to find the causes of a minor stop is to observe it actually happening. Being in exactly the right place at exactly the right time is difficult, however; and even if a minor stop is observed, it often happens so quickly that it is hard to know what went on. This leads people to form conclusions from nothing more than the evidence left behind, with the result that any remedial action tends to be superficial and incomplete. To reduce minor stops effectively, it is essential to spend sufficient time on the shop floor patiently observing, analysing and stratifying them before deciding what their causes are and how to eliminate them.
(4) Eliminating them is an essential precondition for unattended operation When minor stops are rife, the following ill effects occur:
- Automated lines have to be staffed by more people than originally intended in order to achieve the required availability and output, because the original crew cannot cope with all the problems.
- It is impossible to achieve the required ratio of machines to operators (for example, despite it being theoretically possible for one operator to look after 15 machines, in practice 10 is the limit).
- Little is gained from running the line unattended during lunch breaks, because it stops producing after the first minor stop.
- The level of defectives and rework stays high.
Many production areas aspire to unattended operation, hoping to progress from attended operation (where materials still have to be supplied and removed manually) to partly unattended operation (where the line runs on its own during lunch breaks and other limited periods) and finally to fully unattended operation (where the line can run on its own all night). Various engineering and technical solutions are implemented at each stage, but the problems of minor stops and cutting-tool life often remain unsolved.
Many lines that have reached a high level of automation, with fully-automated processing equipment, materials handling equipment and measurement systems, still suffer from frequent minor stops. As a result, the hopes that a high level of automation would permit the lines to be operated without anyone present during breaks or between shifts are not realised. Even when a line has a high Cp value (i.e. no quality problems) and never breaks down, little will be gained from unattended operation if a minor stop occurs right after the operators have left, or even half an hour or so later, because the line will then be inactive for the rest of the unattended period. Nevertheless, although many companies experience this problem when they first attempt to run their lines unattended, they usually succeed in reaching their goal by carefully observing the minor stops and finding effective solutions.
The most important issue in achieving unattended operation is thus to look very closely at how to extend the mean time between minor stops. It is an issue that absolutely must be resolved, particularly if night-time unattended operation is the goal.
11.4 Strategies for Reducing Minor Stops
(1) Analyse phenomenon and workpiece behaviour
Although the most important thing in reducing minor stops is to observe them actually happening, the opportunities for doing this are limited, making it difficult to devise precisely-targeted remedial action. Phenomena should be videoed (using high- speed cameras for very fast minor stops) in order to obtain a correct understanding of the mechanism by which they occur. However, observing the workpieces or products even in the absence of a minor stop (their movement, orientation, angle, etc. and whether they bounce, ride up, vibrate and so forth) in order to identify patterns of behaviour is also helpful. Each action of the equipment (the timing and angle of its contact with the product, its speed of movement, shock, vibration, variability in stopping position, etc.) should also be closely observed. Even when the products and the equipment appear to be moving in the same way all the time, close observation will reveal subtle patterns that can shed light on the mechanism behind the phenomenon.
Observations like these should thus be used to elucidate the mechanism by which the minor-stop phenomenon occurs – theoretical explanations should be found for why the products move as they do, while subtle deficiencies in the way the machine performs each of its actions should be identified, and possible causes of those deficiencies in terms of the way the machine is constructed should be investigated.
To summarise, the minor stop phenomenon, the behaviour of the products, and the individual movements of the equipment should be observed in order to understand the mechanism producing a minor stop. Careful observation of the products and the equipment often enables the causes to be identified.
(2) Correct slight defects
Slight equipment defects are such a common cause of minor stops that the first thing to do in reducing the latter is to find and correct every slight defect on every part of the equipment that could possibly come into contact with the product. In some cases, a single slight defect causes a minor stop, while in others, a number of them act together. For example, when parts are supplied to a machine by an automatic parts feeder, there may be a very slight misalignment between the bowl and the chute, creating a small step at the point where the chute joins the bowl. Normally, this might not be a problem, but if some of the parts have small burrs on them, one of these could catch on the step, jamming the chute and causing a minor stop.
Spotting slight equipment defects entails detecting the subtlest of surface blemishes, which may require magnification, the development of new measurement techniques, sharpening the senses of the operators, or otherwise increasing the precision of observation and analysis. It is particularly important to establish clear standards of judgment for slight defects that cannot be quantified, when the human senses afford the only means of discriminating between what is acceptable and what is not.
The following approaches should be used when searching for slight equipment defects, because they will only be detected if the equipment and the materials that it processes are looked at in a new and different light:
- Search with an awareness of the problem in mind
- Compare the equipment in its present state with what it would be like if it were in perfect condition
- Treat as a defect anything that could reasonably give rise to suspicion
Finding and correcting slight equipment defects in the way described above can halve the rate of occurrence of minor stops and eliminate them from certain parts of the machine. This is why it is so important to spot and deal with even the tiniest blemish (including ones that are so slight as to make it very difficult to decide whether they could be called a defect or not) on any part of the equipment that could come into contact with the product.
The purpose behind correcting slight equipment defects is to minimise minor stops whose location and rate of occurrence change from day to day or from lot to lot. It reduces the number of possible causes, and forces the minor stops to appear in a different guise (see Tables 4.18 and 4.19).
(3) Ensure that basic conditions are observed
In many cases, minor stops occur because basic workplace conditions (cleaning, lubricating and tightening) are not properly observed. Minor stops are inevitable if the equipment is never cleaned, is not properly lubricated, or slackness, play and vibration are left unchecked. Sustaining basic conditions must be integrated into the workplace culture in order to prevent this.
(4) Inculcate standard working practices
The rate of occurrence of minor stops is often affected by the way in which changeovers (including adjustments) are performed. Even the same operator can produce different results on different days depending on the quality of his or her work. This means that the methods used to set up and adjust the machines must be reviewed to ensure that they are correct; all too often, operators forget to do things the way they have been taught, or believe they know a better way of doing it. A thorough review should be implemented to ensure that this does not happen.
Eliminating slight equipment defects and ensuring that basic equipment conditions and standard working practices are observed will alter the nature, location and frequency of minor stops but will not usually eliminate them. Further reductions require a deep analysis of the phenomenon.
(5) Optimise conditions
Before thinking about modifying equipment, jigs or tools, the processing conditions and the conditions under which the components and units have been installed should be reviewed. Processing conditions are physical characteristics such as air pressure, vacuum pressure, amplitude, feed rate and so on, while installation conditions include positions, angles, and resonance. Existing conditions will often simply have been set by extrapolation from those used in the past and may not necessarily be the best. They should be carefully reviewed, and the optimal conditions determined by trial and error or from the results of experiments and tests.
(6) Eliminate weaknesses
If minor stops fail to decrease even after the strategies described above have been applied, it is usually because there are problems with the design of the equipment, jigs, tools or detection systems used; for example, minor stops often occur when jigs are mismatched to the parts they are supposed to hold. Sometimes, minor stops occur because weaknesses in the design of the equipment are not recognised, or because the design of the equipment makes it very difficult for the operators to adjust it correctly. In cases like this, the design weaknesses must be accurately identified and corrective action taken. Generally speaking, design problems are unique to individual products rather than applying to a range of different ones.
It is important to follow the preliminary steps of correcting slight defects, ensuring that basic conditions are observed and so on before modifying the equipment’s design. It is a mistake to start with design modification, because problems are not usually due to design weaknesses on their own but rather to a combination of these with other factors that could be eliminated by restoring the equipment. If not removed, these other factors obscure the problem, making it impossible to identify the design weakness precisely.
Table 4.20 shows the procedure by which minor stops should be tackled.
Chapter 4. Focused Improvement. Part 4
Equipment breaks down every day and lowers production OEE, to avoid or reduce (eliminate) …