TPM

In this submission I shall create a template to illustrate how a TPM (total productive maintenance) program can be implemented in a company manufacturing guitar bodies and necks to both the trade and the general public, describe some of the tools that can be used in the TPM process, and show by example how these can be used.

The company has been operating for twenty years and during this time have expanded from a two man operation producing twenty components per week, to a mediums size operation employing one hundred and twenty people, and producing four thousand units per week. The core business involves manufacturing replacement guitar necks and bodies. The manufacturing process uses a variety of woodworking equipment, including 5-axis CNC routers that are capable of producing parts that require very minor, if any, additional finishing before despatch to the customer. There are however many processes that have been retained as the company has expanded that are labour intensive. The manufacturing processes are in essence:

§  The wood stock is dried in kilns

§  The wood is cut into blanks for machining

§  The blanks are fitted to the CNC machine and cut

§  Or  the blanks undergo a three stage process

§  The finished product is inspected

§  Rework is carried out if required

§  The products are placed in the stock racks

§  The products are inspected, packed and despatched.

The original three stage process is retained as it allows the company to continue produce high value “one off” custom products specified by the customer. It also provides extra capacity it time of high demand for the standard products, and to carry out rework on defective products produced using the 5-axis CNC machines. The programming and use of the 5-axis machine for these custom products would not be economically viable.

The company organigram shows how the company is structured:

 

 

 

 

 

To implement the TPM this exercise will follow the “Eight Pillars of TPM” also known as the “Eight Pillars of Japanese TPM” which are:

§  Initial Phase Management

§  Quality Maintenance

§  Autonomous Maintenance

§  Planned Maintenance

§  Education and Training

§  Environment, Health and Safety

§  Admin Support Systems

§  Focussed improvement

 

Initial Phase Management

The Initial Phase Management covers the introduction and inception of TPM. “It is the organization and planning pillar of TPM” (Borris).

 Changes in the way a company operates, and particularly major changes, are often viewed with suspicion by employees. It is vital at the initial stage of TPM implementation that staffs concerns are allayed; otherwise the implementation of TPM will be very difficult. For the success of a TPM initiative to be maximised it is important that all employees are included, it is also important that it has the support of management, that it is a management driven initiative. Our TPM team at this stage will consist of the Managing Director, Factory Manager, Production Manager, Sales Manager, and Sales Manager. In a larger organisation the team structure may be different, but in this size of company these five managers should have between them a thorough understanding of how the company operates, and be aware of the company’s goals and objectives. Together they shall examine all areas of the company, examining in detail the processes involved in supply of the raw material, product manufacturing, maintenance, sales and customer service, and administration. The reason for making improvements in these areas is to increase profitability by improving quality, reducing waste, and most important of all, to increase customer satisfaction levels. Prior to commencing TPM and company meeting is held to explain to employees why the company needs to introduce TPM, outline the expected contribution from the employees, and explain the benefits to both the company and the employees.

In this case study the production and maintenance functions shall be examined in detail. It is important to note that all other functions must be considered to ensure the maximum benefits are achieved, for example the admin function must liaise effectively with production to ensure the raw materials are available when required, are of the expected quality, and that the stock levels are optimised .

The TPM team is aware that a significant amount of the products being produced either require labour intensive rework, or end up being scrapped.  This is an area of concern that up to now has not been examined in enough detail, but is an obvious area that would benefit from “focussed improvement”. Focussed improvement, simply described, is where attention is focussed on a particular function or process to improve the overall equipment effectiveness (OEE). This can be achieved by improving the performance of the equipment, the staff, and quality. All contributory factors reducing the OEE are losses. The aim of focussed improvement is to eliminate losses.

As part of the greater project team activities, the Production and Maintenance Manager shall, with the help of other employees form a team, with a common set of goals, to examine the processes. This is one the fundamental changes that TPM requires, no longer is it a case of production and maintenance (and other departments) having different goals, instead it is the disciplines acting as one focussed unit with a common goal. The common goal in this case is to improve the manufacturing processes, and drive the business toward the goal of zero losses.

The first step the TPM workgroup take as part of the improvement plan is the measurement cycle. This is essential as any subsequent changes to the processes must be measurable. Many of the changes made may be small incremental steps, and it important that any benefits can be identified and assessed.  

Following the initial examination of the manufacturing and maintenance functions the following data has been captured for the eight CNC machines that produce the majority of the parts. The data considers the components produced by the CNC machines over a 4 week period. The CNC machines made by three different manufacturers and most have been modified in some way by maintenance.

 

Equipment

Guitar Bodies Produced

Guitar Necks Produced

Requiring rework

Scrapped

Downtime

CNC  1

376

1102

76

5

1 day

CNC 2

564

534

5

87

15 days

CNC 3

49

1287

134

84

4 days

CNC 4

0

2358

6

8

0

CNC 5

1877

338

273

103

0

CNC 6

340

786

39

22

2 days

CNC 7

0

2123

23

2

0

CNC 8

0

2432

134

36

0

 

It is clear from the data that there are a great number of defects, the reasons may be varied, and it is almost certainly the case that the operators and maintenance are aware of the reasons behind these failures, but in order to address these issues they must be identified and prioritised.

One technique our team may use to indentify the failure causes is to conduct a brainstorming exercise. This requires the participation of the employees who have knowledge of the systems. This could include operators and maintenance technicians. The brainstorming exercise is usually conducted over a fixed time, for example 30 minutes, and everyone is encouraged to suggest causes, no matter how obvious or unlikely they appear, and they are all noted.

A useful tool to employ at this stage is the cause and effect analysis.  One way of performing this is using the Ishikawa or fishbone diagram. In this case using the suggestions from the brainstorming exercise we may end up with a diagram like this.

 

 

The green boxes are affinities, the blue boxes are causes, and the red box is the effect. The affinities in this case are the two main product groups, guitar necks and guitar bodies. If the blunt router bit cause is considered, it appears to have two associated effects, wood split and rough finish. These are not effects however they are causes. A rough finish or wood split mean the product must be scrapped or rework carried out, which is the effect.

It is clear from the fishbone diagram that the data we have is not detailed enough, as it only lists the number of failures; the fishbone analysis gives the main perceived causes for the failures. This data is still useful however in determining where to start the detailed measurement phase. CNC5 produces the highest amount of defects at just fewer than 17% of parts produced, so that will be tackled first. In order to carry out our focussed improvement we have to ascertain the extent of the problems. By collecting further data on CNC5 we capture the following information.

 

Failures

Body rework

Body scrapped

Neck rework

Neck Scrapped

Machine not suited to task

12

6

              0

             0

Marks from debris on machine

11

0

              5

0

Wood split

0

11

0

0

Rough finish

74

6

82

58

Machine stopping

0

0

0

0

Blank not positioned correctly

0

0

0

12

 

It is sometimes useful to show the results graphically, and one method is to use the Pareto Chart. The chart is organised with the most significant data to the left, the remaining data is display in descending order with any failure or combination of failures totalling less than 5% grouped under others. This technique clearly shows where the main type of failure is occurring.

Pareto Chart of Defects.bmp

It becomes clear that most of the problems are as a result of the tooling failing (blunt router bit), and although all the failures must be addressed to achieve the goal of zero failures, concentrating on this particular one first, we know that the tool is failing but we don’t know the reason. The production reason may be that the maintenance department are not replacing the tools when required; the maintenance department may say that the production won’t release the machine. In order to find out exactly what’s causing this particular problem we can use a technique such as the “5 Whys” or “asking why five times”. The 5 Whys is a technique used to drill down to the root cause of the problem; it involves asking why, the required number of times until the root cause is found, not necessarily five times. This technique must be explained in advance, as the people being asked may feel their competence is being questioned. It may yield results such as:

Question             Why are the router bits becoming blunt?

Answer                Because maintenance are not changing them when asked

Question             Why is maintenance not changing them when asked?

Answer                Because they are busy attending to breakdowns

Question             Why are they busy attending to breakdowns?

Answer                                Because the are behind with their preventative maintenance

Questions           Why are they behind with their preventative maintenance?

Answer                                Because they are spending too much time on tool changes

 

In just four questions we have identified the problem, and the possible solution is also quite clear. The problem is maintenance has too great a work load, and the obvious traditional solution is to take on more maintenance staff. But is that the best solution? That will actually increase costs, not reduce them. The 5 whys can of course be applied to other problems, including equipment failures.

 

One of the cornerstones of TPM is the concept of autonomous maintenance. In this case the operator who is familiar with the machine knows when the router bit needs changed. The skilled technician has to schedule this into other maintenance tasks, and while this is being organised the failure rate is rising as the production operator strives to meet targets using inefficient tooling. The solution here is to train the operators to change the router bits. In the interests of safety (another pillar of TPM) this must be done correctly, to ensure the operator is fully competent. This could involve the maintenance performing the initial training, producing a clear procedure, and supervising the operator initially to ensure that the task and any associated risks are fully understood. A detailed task risk assessment would be conducted to identify the risks. A single point lesson (step by step instructions) for the task could be produced and sited in a suitable position on the machine. By devolving this task, and other minor maintenance and repair tasks that are to the operator valuable maintenance time is freed up to allow the maintenance resource to be focused on tasks requiring the greater technical skill levels, such as RCM analysis, delivering more effective preventative maintenance, and perhaps leading to beneficial design changes.

 

One of the “eight pillars of TPM” is education and training. In this particular case, following suitable training and competency assessment, the role of trainer could devolved to one or more of the production operators. This can have beneficial effects, not just for the company, but for employees. It can provide motivation for employees who feel their talents are being underutilised or that their job is “dead end”.

 

The second most significant cause of failures is “marks from debris on machine” this tends to suggest a degree of worksite untidiness or disorganisation.  A technique called 5S or CANDO can be used to address these problems. The technique originates from Japan and its function is to ensure a clean, tidy, and organised work place. The “5S’s” come from the Japanese words Seiri, Seiton, Seiso,Seiketsu, and Shituke. These terms have been translated into English and rearranged to CANDO as shown in the following table, though the original order is more logical. 

 

Seiri

Sort

Arrangement

Cleaning

Seiton

Straighten

Neatness

Arrangement

Seiso

Shine

Cleaning

Neatness

Seiketsu

Standardise

Order

Discipline

Shitsuke

Sustain

Discipline

Order

 

 

Applying 5S to our CNC shop floor we could first consider “Seiri” or arrangement. First the plant is divided into zones. For a zone containing four of the CNC machines as shown we may consider the following:

 

 

·         Are the CNC machines sited in the best positions?

·         Is the Woodstock (blanks) stored in the ideal location?

·         Is the finished product stored easily and in a way it cannot be damaged or marked?

·         Are there any items or equipment that is no longer used?

·         Is the area clean?

 

 

 

Text Box: Wood Stock Text Box: CNC1 Text Box: CNC2 Text Box: Product
 

 

 

 


Text Box: CNC3Text Box: CNC4 

 

 

 

 

 

In the original arrangement there are two problem areas. The first is the risk of wood debris (chips) from one CNC machine flying onto another CNC machine as the blank is placed, causing a defect on that product. The second is that CNC 2 and CNC4 are further away than necessary from the wood stock storage area, and CNC 1 and CNC3 are likewise away from the product storage area.  The first problem may require a design change, perhaps Perspex guards on the machines, but the second can be improved simply by changing the arrangement of the machines as shown. By focussing on this area improvements can be achieved that may otherwise have not been considered. The position of the machines although subject to consideration when installed, may not be optimal. Without this focussed improvement effort it is most probable that “workarounds” would be implemented to overcome any problems caused by the positioning of the machines, and these would become “normal practise”.

 

Text Box: Product
Text Box: Wood Stock Text Box: Product
Text Box: Product
Text Box: Product
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 


The second “S” is Seiton or neatness and in the above example has gone someway in addressing this, but other areas may be considered such as the type of shelving used to store the wood stock and finished product, to ensure everything is organised correctly, wood stock and finished product are easily accessible, and that they are not damaged.

The third “S” is Seiso or cleaning.  The responsibility for cleaning must be clearly defined. In the above example it may be decided, not unreasonably, that the four operators of the CNC machines would be responsible for the areas around their machines, as well as jointly for the cleaning of the wood stock area. It may be necessary to have set inspection levels to ensure this is done.

The fourth “S” is Seiketsu or order. Once the acceptable standards have been achieved these can be maintained with perhaps a picture of the “ideal” area posted appropriately as a reminder of what standard is expected.

The fifth “S” is Shitsuke or discipline. There is no point in making improvements as a short term measure. In order for the benefits of the exercise to be fully effective they must be maintained, or indeed improved.

The “5S’s” are simply a way of structuring what is common sense, in a way that it can be effectively applied in an industrial environment. The lines between the 5S’s are not well defined. There is a great deal of overlap and ambiguity between them, but the result is that all the potential problem areas that may undermine the exercise are addressed.

The final of the pillars being considered is “Planned Maintenance” also known as “Preventative Maintenance”.  If correctly applied planned maintenance can be very effective, if incorrectly applied it can be a source of waste, either by virtue of the fact that it does not improve reliability, or indeed reduces it.

Reliability Centred Maintenance (RCM) is one method used to assess the requirement for preventative maintenance. Basically the functions (what is expected) of the equipment being assessed are identified, the functional failures (in what way it fails to meet expectations), and the failure modes (how it fails) both known and potential are identified and these are noted. The consequences of these failures are then analysed and from this a decision can be taken on whether planned maintenance is worthwhile, and if so how frequently. It can also determine whether there are environmental or safety issues that make this maintenance mandatory, or may even demand a design change of the equipment if this is not technically feasible. RCM also considers predictive maintenance as a way to determine how much maintenance is required, and when. If for example equipment has a known PF interval, where P is the point where it starts to fail, and F is where it fails, then on condition maintenance can be carried out to restore the equipment when it is clear that the maintenance is actually required. There is little point in expending valuable maintenance resources trying to restore equipment to a condition it is in already. “Wrong maintenance eats production time and bad maintenance has a banquet” (Borris). Although TPM and RCM are different tools, RCM can be used as an effective component of the TPM approach. For RCM to be fully effective the equipment should be in base condition, which is it should be either new, or fully restored.

One TPM technique that is used to improve the effectiveness of maintenance is PM mapping. One of the major advantages of this technique is that is considers the equipment in the context of its current use. An example of how this may be used is shown below. In PM mapping as with other TPM techniques, the information is presented visually. Using historical data a malfunction map is created using the data to identify all the failure modes. By using coloured dots to represent the faults, information on all the failure modes can easily be conveyed, and a suitable maintenance strategy can be developed, or the existing maintenance improved using autonomous maintenance, and more effective use of the maintenance function:

 

                White F-tag showing faults that can be fixed by autonomous maintenance

                Black F-tag showing minor faults not requiring a machine shut down

                Red F-tag showing failures

                Yellow F-tag showing past failures form equipment history.

               

The coloured tags are simple attached to the image of the equipment. Where there is a high density of tags then the solution is simply to enlarge that part of the image to show more detail.

Router.bmp

Once all the failure modes have been identified a PM map can be created for the equipment using a similar technique as for the malfunction map. In this case blue dots are used to show what part of the machine has existing PM routines, and green dots to show failures. For this to be of any use the equipment must first be brought back to base condition. In this condition there will only be blue dots to represent the current maintenance, the green dots would appear as failures occur. This quickly shows how effective the maintenance is, ideally there should be no green dots showing that the maintenance is effectively preventing any failures. No green dots of course may also mean that too much maintenance time is being spent, so the inspection intervals must be reviewed to get to the ideal levels. The use of visuals for the malfunction and PM maps is of course only part of the picture, the normal data collection and recording (equipment history) is still required.

 

 

 


In conclusion this exercise although only scratching the surface of TPM, does touch on all of the “eight pillars”. It also shows that in today’s industrial environment there is no longer a place for the old working practices with their roots in Taylorism. In order for any company to remain competitive in the Global Economy it must use not only invest in the most efficient equipment, it must also utilise the talents of the workforce to achieve its objectives by effective training and education. It must ensure that the existing equipment, human resources, and processes are optimised. The tools and techniques available to achieve this are many and varied. If used appropriately they can enable a company to achieve the fundamental objective of TPM, to reduce losses to zero.

 

 

References

Borris, Steven. (2006) Total Productive Maintenance. New York:McGraw-Hill.

ISBN 0-07-146733-5

 

Moubray, John. (1997) Reliability-centred maintenance: 2nd ed. New York: Industrial Press.

ISBN 0-8311-3078-4

 

Smith, Steve.(1997) Solve That Problem. London: Kogan Page

ISBN 0-7494-2482-6

 

Acknowledgement

 

CNC Router Store for images of 5-axis router

http://www.cncrouterstore.com/detailedinfo-5axisrouters-8471.html/[accessed 31 Dec 2007]