The Michael Ford Manufacturing Blog

The Michael Ford Manufacturing Blog RSS

Where Is The Manufacturing World Coming To?

May 16th, 2013, by | Permalink | No Comments

3D printing technology is really growing in popularity now. Having worked through the growth of sophistication in electronic product manufacturing, I can see how this technology is likely to develop. Today, most printers use a single nozzle with a nylon based plastic paste that is effectively dispensed like an SMT glue machine, except, that layers are built up over each other. It really is that simple. If the technology follows a similar path to SMT, we can expect to see multiple nozzles in the future, each dispensing a different compound, perhaps we will see multi-head machines, even those that can change the nozzles as they go, and even modular machines. The dispense “elements” will of course include different textures, strengths and colours of plastic, but, also you could imagine these could be conductive elements, such as embedded electronic components. It will be quite a while I expect before we can print off something like a working cell-phone, but, you never know……

Such kind of manufacturing would completely change the way the manufacturing industry works. If everyone had a printer, then we simply buy the supplies, in the same way as ink cartridges (you know how expensive they will be!) and then simply purchase the design of whatever we want to build and own. Traditional factories could become a thing of the past, and with them, distribution chains.

Perhaps this technology is quite timely, as there is an increasing amount of discussion now about the costs and risks associated with remote manufacturing. It is a complicated issue. Anyone will understand that factories operating nearer to the main consumer areas in Western Europe and the US would be far more expensive than those operating in lower-cost areas such as South America, China and many parts of Asia, both in terms of labour and also general operating costs. On the other side of the balance however are the costs of remote manufacturing. The transportation costs are the most obvious, which continue to increase rapidly. There are other costs however, associated with the time delay from leaving manufacturing to point of sale, which delays new products into the market and also builds a much larger than necessary inventory of stock in the distribution chain which can then be affected by significant price erosion, especially as products near end of life. Remote manufacturing operations also bring risk, not least from the weather, earthquakes, potential shipping problems, and lack of control of quality and product content. There is increasing pressure now from the side of the cost of remote operations, causing people to look again at the options of local manufacturing. Other than the cost issue however, there are other issues with the local manufacturing option The first is with materials, since the manufacture of materials has followed the assembly operations to the remote locations. It will take some time to bring these back, and until then, materials will have to themselves be imported from the very same remote areas. The other issue is with skills. There has not been a great call for manufacturing engineers on the actual shop-floor recently in Western Europe and the US. As people with skills have retired or moved to other opportunities, often, they have not been replaced. Even those going into consultancy have recently seen opportunities diminish.

Manufacturing in Western Europe and the US will not however want to return to what it once was. It needs to be so much smarter, agile and flexible with no tolerance for ineffective operations, and will experience a steep learning curve to do so. This is where specialist software systems focussed on SMT and electronics product production come in, bringing with them the latest innovations and world-class operation flows. Though we cannot pretend that  Mentor’s Valor MSS Suite will make returning and restarting manufacturing operations child’s play, we think that it will provide the best head-start an operation could get. Just as well you cannot use a 3D printer to make software…….

Tags: , , , , , , , , ,

Intelligence? More Like Complete Stupidity…

May 1st, 2013, by | Permalink | No Comments

Unbelievable! Don’t get me wrong, it is not the fact that pop-up ads appear all over commercial websites that annoys me; after all, we have had the same thing in magazines and newspapers for many years. Advertisements appear almost everywhere these days, the funding from which is an important part of many companies’ business plans. This is not the issue. The issue starts with the difference between physical ads and on-line ads. On-line ads can be dynamic; they are now tailored for the specific individual. The hype has been around the intelligence to target specific products and services that the person is interested in, thereby getting a higher success rate to that person responding to the ads. This is put forward as a form of “intelligent advertising”.

In my experience to date, this is complete garbage. I go on-line to buy an external hard disk, do a few searches to find the model that I want, then a few searches again to find the best price, and then, I buy it. Great. What happens next? Loads of “intelligent” targeted advertising telling me about external hard disks. The same ads appear on pretty much every commercial website that I go to. Seriously? This is the sum total of the advertising intelligence? Just find out what the recent searches were made, and then inundate the person with ads for similar products? Having just bought an external hard drive, what are the odds that I would be now looking to buy another one? Same story for the USB stick, the laptop, the office chair, the greenhouse. The fact that a purchase was made is a very critical piece of information that obviously is not recorded and results in the “intelligent” advertising being the dumbest of all. Someone, somewhere, thinks that this is acceptable, perhaps with the justification that people go on-line to do shopping not to actually buy anything most of the time. Really?

The intelligent advertising story sounds good, until you start to think it through. The missing piece of the purchase event illustrates how far away this “intelligence” is from the way that it was originally portrayed. The story was that it is supposed to “know” the person, to be able to understand their wants and needs, to be able to identify products that would be of interest. I was expecting that the intelligent advertising would show me external hard disks before I thought I needed one. A big gap between perception and reality.

I may just be over-reacting, I guess that I have grown sceptical over the years. I really hate it though when people bring “solutions” that are at best just “box ticking” exercises where the story sounds good superficially, and at worst, just a way to make business at the expense of the customer, no matter what.

I know that there are other people who have been “burned” by a solution that had sounded good, where the expectation and the reality were quite a way apart. There has been a wide gamut of solutions out there for PCB-A production, each making their own pitch about what they can do. How to avoid the “intelligent advertising” effect? There is great value in working with a partner who knows the industry, knows your situation, knows the pain points that you are feeling, and in fact, knows what will be the major issues a year or even five years from now. Anyone though can say this. The proof has to be there, demonstrated and proven by the introduction of innovation, solutions that were conceived years before the industry realised that they are a must in order to remain competitive, but now are available when needed, are mature, and really work. Just in time material management, simultaneous work-order and feeder commonality optimisation for production planning, just two of the latest hit technologies, originally conceived years ago, and today available to really meet the requirements of electronics production looking to get ahead, to differentiate themselves. Beware of crude imitations….

Tags: , , ,

Nepcon China – The Chinese Way

April 5th, 2013, by | Permalink | No Comments

Nepcon China Rocks! There is so much going on, so much potential, so many companies focussed on success, to be the best in their field, to be number one. Electronics in the West is like the Mississippi river, huge and powerful, but slow moving, quite boring actually. In China, it is like riding the rapids. This is where skill and vision come in to play.

Creating the best production factory is the key, a factory that can make anything, at a moment’s notice, in an efficient and reliable way. Experience and know-how are important, to know what is right and not to cling to the momentum of obsolete practices that resist innovation to change. The SMT shop-floor is complex, comprising several different key roles, each with their own priorities and flows, each needing to work together for the operation to work optimally.

This is 2013, software systems are already all over the shop-floor, often as islands designed to, for example, operate or speed up a specific process, to monitor the performance of key machines, to gather defect information from repair stations. These kinds of “islands” of software have served the industry well for some time, growing up as in-house solutions, others from local suppliers who have made a business from developing solutions. Moving forward, how can software systems be used to differentiate the business against the competitors? Competitors after all, will also probably have similar systems and experience. To break out of the crowd is the goal, to create that clear differentiation.

As children grow up, they need bigger clothes. It is the same with software. Systems that offer point solutions, that is applications developed bearing in mind a specific issue and need on the shop-floor, are soon outgrown. As the business needs evolve, objectives change, the systems also need to change. Having many systems on the shop-floor has also no doubt led to the duplication of support effort and overlap of functionality. The next step must include having solutions that work together, sharing a common infrastructure, flexible enough to cope with any production innovation that comes along. In fact, these systems themselves should bring innovation into the operation. Whether it is new product introduction in half the time, optimisation of SMT for high mix and short notice changes whilst maintaining high efficiency, or whether it is control of materials to reduce costs and gain traceability, these are the opportunities to differentiate the business through innovation that competition does not yet have.

Nepcon China in Shanghai on the 23rd to 25th of April, should be a good show. For those attending the show, please don’t miss the opportunity to come to our Mentor Graphics booth 1A18 and talk to us about how we can help you to differentiate your business with our Valor MSS Software Suite.

Tags: , , , , , , , , , , , ,

Stop Horsing Around – The Need For Traceability

March 21st, 2013, by | Permalink | No Comments

They say that you are what you eat. If so, it means that many people in the UK are rather more horse than they thought. Endless jokes about, “why such a long face?” aside, this really shouldn’t have been a surprise. For years now, there has been food on sale in the UK simply described as, “meat”, such as meat pies, meat sausages etc. When someone asked about what the meat actually was, DNA testing was done. A variety of different animal contributions were found, including a significant amount of content which could not even be identified….. The real anger has rather been more down to the fact that horse meat was found in products labelled for example, as beef. The UK shop-going public hates to pay premium prices for what is a cheap alternative.

This issue illustrates that something that is relatively stable can quickly “blow up” into an issue as the unknown becomes exposed. Electronics manufacturing faces the same issues. We see presentations at trade shows that focus on the ever increasing problems of counterfeit materials. These can be reclaimed, salvaged, second use materials, even some which may have had leads replaced or tinned to disguise the lead content, as well as just low quality copies.

With the horse meat “scandal”, there was another sting in the tail. Horses apparently can be treated with drugs which are potentially harmful to humans. Straight-up horse meat in the food chain is free from these chemicals, but no-one can be sure about the illicit meat. Low cost electronic components may work perfectly well in low cost devices or products designed to be built using such components, but how about the safety systems in your car or medical equipment. How about the life and quality of your premium branded products, even the lower cost models? Can the risk be taken?

Many years ago, production operations had the scope to do random material testing, going into quite some detail as to the content and reliability of even passive components. Today, few operations are doing this, having put their trust into the supplier relationships. As times get hard, relationships strain, local alternatives are found, the testing process may not be replaced.

Material traceability is therefore of critical importance for production. It is the ability to link a materials related quality issue, whether discovered during the production operation or whether found as a defect in the field, back to a specific supplier, vendor and batch of components. It is the ability to immediately identify the scope and likely effect, in a business sense, of any extraordinary materials being used. Just the knowledge that material traceability is being used effectively, such as with the Valor MSS Materials solution, can be enough to prevent unscrupulous suppliers trying to get away with the supply of less than optimum materials.

It is therefore a surprise to hear from machine vendors who tell us about the buying patterns of companies buying new machines. The answer to the question of whether they would like verification or traceability is usually a “neigh”. Perhaps the food is getting to them…….

Tags: , , , , , , , ,

Address Your Challenges And Gain Your ROI Within 9 Months

February 13th, 2013, by | Permalink | No Comments

The World-Class Executive’s Five Year Survival Plan

Throughout successive peaks and troughs of the business climate, there are always opportunities to succeed over competitors in the market. The more volatile the electronics industry becomes, the greater the opportunities, but the more difficult it may seem to address them. Today, with business being challenged in so many areas, and with such key changes in the ways that the electronics product market is working, there are some amazing opportunities for those who are prepared to think and act, perhaps some would say, more radically than in previous times. Some core behaviours and attributes of the production operation may need to change, challenging the core of what has been put in place, never something to be taken lightly.

The world-class executive is expected to take the responsibility for making decisions which follow a vision to steer their business through to success, for share-holders, employees and for that personal satisfaction. The five year plan is the essential tool to explain this vision and put key milestones into place. It is not a 5 year plan where success comes at the end; it should be a plan of continual success throughout the period, with core changes introduced at appropriate times.

At the APEX show in San Diego next week (19th – 21st February), as well as making presentations at our Mentor Graphics Booth (#1227) on this and related subjects, I will also be presenting a poster which explores the areas that the world-class executive in electronics manufacturing should be including in their plans, looking in detail as to how the industry and especially the way in which electronics positions are sold in the market, now drive the need for manufacturing to bear the brunt of a great deal more demand for flexibility according to unpredicted demand pattern from the customers. Removing the barriers and costs for the production operation to be agile is therefore a key issue. Some barriers are physical, such as the overhead of managing large stocks of materials that accumulate on the shop-floor. Others are bottlenecks of engineering and planning. Engineering, planning and above all, materials, are key silos in the production operation, each containing a lot of history and momentum for the way they work. This is a good thing much of the time as small errors can have catastrophic effects to the business performance. Sometimes however, things need to move on, MRP technologies are still based on 1970s derived ideals. Changes made however need to be practical and real, so messing about with the technology of the day or meaningless boxes that would-be solution providers bring along to tick, is definitely not the way forward. Core improvements need to be made in a managed way by people who understand the reality of these technologies, with experience over many years.

I invite you all to come along to the APEX show, visit our booth, come to see the theatre presentations, come and see the poster presentation. All challenges welcome!

The poster presentation is scheduled for 3:30pm on Wednesday 20th, and is free (please pre-register at www.ipcapexexpo.org/register.)

The theatre presentation schedule at the booth #1227 (just walk up in time for the start of any presentation) is:
 

Tags: , , , , , , , , , , , ,

Extreme Chicken & Egg

February 6th, 2013, by | Permalink | No Comments

It is always interesting to hear about those companies who have actually taken radical action to succeed, rather than just talking about it. Seeing changes in market demand and hearing how new tools can help, they jump in and address the challenges head-on. Making the changes creates for these companies something that in a real business sense, sets them apart from their peers.

Such a success story I found recently in Germany, where generally business is booming as compared to the rest of the world. Industry leaders in this leading country have embraced agility. Agility, or more precisely, the lack of it, is one of the key ways in which we can measure the abilities of a production company. For almost every manufacturing operation, the high mix of products and decreasing dependability of the sales forecast means that more and more frequent changes have to be made in the production operation. The old guard of manufacturing people will always tend to assume that higher mix and higher volatility are the enemy, as they expect it will always result in poor performance, such has been their experience.

This simply is not true anymore however. My colleague, Mark Laing, put together a webinar recently where we talked about  “Extreme Planning”. It talked about the optimisation of materials loading on machines at changeover between products running on SMT lines, and also about the optimisation of work-orders to meet customer demand. Both of these things are not new, they are done pretty much in every manufacturing environment by some system or other, even if they have to be done manually. One question that came up at the end of the webinar was, “What makes Mentor’s Production Planning Extreme?” I had chosen the word “extreme” for the title of the webinar with the intent of setting the expectation of taking SMT planning to the next level, where perhaps it had been thought in the past unattainable in reality.

Planning for SMT processes has been a long term issue in the market, where there are many standard tools to choose from, and yet none of them can provide a satisfactory solution. It is a situation of chicken and egg, which represents the two things that make up true SMT planning. One is to group feeders of different products so as to reduce the time it takes to changeover from one product to the next. This is usually called grouping, with tools available with most SMT programming systems today including those from machine vendors. The issue though is selecting the right products to group, which not only is dictated by the commonality of the products, but also by the sequence of the work-orders to meet customer demand as effectively as possible. Unfortunately, this is the other element of SMT planning. The sequence of work-orders can easily be decided by many tools on the market, but, the decisions are based on machine throughput and changeover times. The situation then will go in one of two ways, whichever chicken or egg comes first. Either the operation will create the work-order sequence and then create the materials grouping according to the sequence, or, the material grouping is made first which in turn drives the work-order sequence. Either way, only a small fraction of possible optimization solutions are considered, many assumptions to whichever the second stage is, are inherited. This will almost never result in an effective operation. The “extreme” element then is to have a tool such as the Valor MSS Production Planning which can do both of these optimisations simultaneously. Only in this way can every possible solution be explored without any artificial assumptions.

This is surely then a real high-value functionality for SMT production operations who have started to see sudden changes in demands and face the needs to continuously re-plan in order to meet demand and avoid making excess stock, especially when you can throw literally thousands of products in to the mix and it takes just a few minutes to crunch the numbers. Extreme? I would say so, extreme value!

Tags: , , , , , , , ,

Fix The Problems, Not The Symptoms!

January 8th, 2013, by | Permalink | No Comments

Fixing problems can seem hard, there is always a momentum resisting change. Is it better though, to spend time and effort to fix the symptoms of the problem? Quite honestly, no.

The UK had a lot of rain recently, but, the growing season is coming around again and the greenhouse is only partially complete, so the week-end was spent outside digging in the mud, again. The next door neighbour comes along, sarcastic comments expected, but no, the discussion turned to the husband’s hernia, which may or may not have been excavation related, and which had been left untreated for more than 18 months, a victim of health service waiting lists.

Even though the hernia is not necessarily life threatening, it cannot be very comfortable to live with. There are the restrictions of what can be done physically, diet, the pain suffered continuously, the cost and side effects of the drugs prescribed and taken to relieve the pain, the cost of care and visits to and from the health service. This is only the start. There is then the risk of further development of the problem. A non-urgent hernia today can easily become much more serious if it grows bigger or the “flow” gets restricted.

The UK’s National Health Service has at some point decided that all of this is acceptable, rather than simply clearing the backlog of cases. Clearing the backlog would mean more cost, more operating theatre time, more surgeons which clearly they thought was out of budget. Some fundamental facts however were ignored, specifically the growing cost of the care of those waiting for operations and the growing cost and complications of the operations as the hernias become worse and more difficult to operate on over time. There is also the administrative cost managing the list of people who have the condition and their changing circumstances. What about the cost to the government paying benefits to these people who may not be able to continue working?

The choice was made. It could alternatively have sorted people out straight away, got away with minor operations with no significant need or cost of the treatment of symptoms and no losses due to people’s incapacity. The “lean” approach as we know it now would be to understand the number of hernia operations, which must be more or less constant, to create mini-operating theatres dedicated to hernia operations that met the demand, at a much cheaper cost than full “ready for anything” theatres. Reduced cost, reduced pain, where was the down-side? Perhaps too radical a thought at the time.

The mini-operating theatre idea came to me as an analogy of the cell production environment in electronics production. In production, there are also pains, there are symptoms, and there are Band-Aids and bandages put around difficult issues which bring costs and restrictions with what can be achieved operationally.

The challenge for production today is that the world is changing. Long term forecasting is giving way to short term flexibility demand as the point for sale moves ever closer to the factory door. The factory will ultimately need to bear the task of making whatever is wanted whenever it is requested, pretty much responding to raw customer demand. Not good then to have many work-arounds related to pains and symptoms restricting in what can be done, what can be changed. These pains, like my neighbour’s hernia, needs to be addressed at source and symptoms eliminated. Manufacturing needs to be fully fit in order to be successful. The Valor MSS suite provides the opportunity to the factory to address the market challenges by eliminating the legacy constraints with tools such as Process Preparation, Lean Material Management and SMT focussed Production Planning, and part of the Valor MSS suite. If only then this application of Lean thinking could rub-off on the UK’s and other countries’ health services, perhaps our lives would be a whole lot more productive and pain free…

 

Tags: , , , ,

The Forecast Is Killing Me!

December 13th, 2012, by | Permalink | No Comments

Grandfather clocks have always been interesting for me. Big old timepieces, made what appear to be hundreds of years ago, with a slow and somehow imprecise tick-tock.  Looking at the second-hand going around, it is not the precise tick of a modern digital device, the hand drops forward to the next second, then somehow rears itself backwards ready for the next lurch forward. It reminds me of the ageing ERP sales forecasting system. 

Actually, my love of grandfather clocks balances with my dislike for the sales forecast. In the past, perhaps the forecast was relevant. There was very little timely communication possible through the distribution chain that was of value. The work was left to sales and marketing teams to assess the market from their perspective and then feed in the information to the ERP system, which justified its millions of dollars investment by adding up the numbers of inventory in all the various warehouses, lead-times etc., and then came up with a plan for manufacturing. It was rather more simple than people care to admit, it worked in most cases, but not all. New products were a complete marketing minefield, and end-of-life products caused a yo-yo effect of boom and bust as salesmen were fearful of the falling value of redundant stock balanced with the demand created by cut-price offers. I saw one case myself where the end of life major product line was suddenly cut short a year prematurely with the loss of hundreds of manufacturing jobs simply because the cost of the yo-yo demand effect to manufacturing caused more incremental cost than the business was worth. The company never recovered.

Coming forward today, the sales forecast is in even more trouble. Fashion and technology have started to merge. I would not be surprised to see a “Paris fashion week” for handsets in the near future. The consumer demand patterns for fashionable goods are far more random that that say for a washing machine, the type of product that most electronics products were in the past. This is a major change. Please don’t ignore it. People buy fashionable goods with preferences for colours and styles, features and functions. This is not only for hand-sets. Cars are also ordered with these options, and are available to buy on-line and at supermarkets. The key issue is that the point of sale is getting closer to the factory. The opportunities for companies to hold minimal stock between the factory and the on-line store is there, and it carries a huge cost reduction incentive. We are now therefore seeing the emergence of the demand signal directly from the point of sale to the customer at the manufacturing site. This is a big deal, it is what Lean has been telling us to do for some time now. Planning the factory to produce according to the near raw customer demand, without the huge overhead of the complex distribution chain, is a major new paradigm for most operations.

The manufacturing site then, to remain competitive going forward, needs to embrace this change as opposed to fighting it. Traditional tools and flows are only going to strain more and more, it is not going to improve. Embracing flexibility and agility, to go week by week knowing only roughly what is going to be built, and being able to optimise day by day to produce the right products at the right time and still achieve the productivity goals, is the road to success.

Is this all a surprise? No, I am sure that many people reading this will have seen this coming, some specialist areas much sooner than others and will have experienced it already. The nice thing today for those now aware of this opportunity, the tools to manage such agile and flexible environments exist already. We saw it coming too. Lean JIT materials, product portability, optimised planning with grouping, all part of the Valor MSS suite available today, off the shelf. You can’t get this anywhere else. Keep this in mind the next time the forecast changes suddenly……

Tags: , , , , , , , , , , , ,

It’s Not The End Of The World!

December 3rd, 2012, by | Permalink | No Comments

Oh boy, the 21st of December is the end of the world. Again. I think in this case, we can be pretty sure that it is not going to happen. Apparently there were errors in the way that the calculations were done with the Gregorian calendar meaning we probably should already have disappeared.

Data can be quite misleading, as can selective interpretation. It happens all over the place, including on the shop-floor. Systems that collect data can be easily misinterpreted. Sometimes it is accidental, a genuine misunderstanding, or perhaps a case where the data or understanding of the data is not explicit enough. In other cases, the meaning of the data can be manipulated to fit or show a particular pattern from which a story can be told.

The stories about the end of the world have not affected me. I don’t believe them, irrespective of where they come from. We see reports however where people do believe, and make huge changes in their lives as part of the preparation. The damage to such people’s lives is incalculable. The shop-floor by comparison is not going to seem so “life or death”, but relatively equal damage can be done.

 I have heard in my experience so many comments about “flavour of the month” where the focus is on a particular topic that is felt to be important, to the exclusion of all else. A month later, a different flavour. What happened as a result of last month? Sure, some numbers improved. Some other, perhaps ignored numbers, got worse. The flavour changes. Was there really an overall benefit? A year from now, the changes fade away. It is very frustrating. Data can lead us down so many incorrect routes.

 As with the “end of the world”, people are looking for the “truth”. My own area of expertise is the shop-floor, so I can only comment on that. We need to understand the whole picture, how things relate to other things, how things combine to create challenges or opportunities. We need to define what control is, what is in control, and what is out of control. We need to be alerted when things go out of control, and take timely controlled action. We certainly don’t need a dramatic off the wall interpretation of isolated data to kill us all off.

This is a large part of the value of reporting from an integrated system such as our  Valor MSS  product in that the complete visibility of all aspects of the operation is available. No more flavours of the month, no more crazy end of the work scenarios please…….

Tags: , ,

Aegis & DiPlan

November 14th, 2012, by | Permalink | 1 Comment

Here we go again. Another merger of companies with significant overlap of portfolios. In the past when this has happened, as in the case with Technomatix, Unicam, Fabmaster etc., many customers were confused about how any “new solutions” would appear. Would there be a move to one standard system, would the systems co-exist, how would they connect? These were initial questions. The next questions were about how customers would migrate. Customers asked whether their investment would be protected. Would all or at least certain product lines be dropped? Would customers, expecting to make an upward move actually be making a sideways step or even a down-grade as I heard from some people’s opinions? It can all lead to a “poisoning” of the market, bringing uncertainty and doubt.

Where acquisitions are done where there is little significant functional overlap, there are far fewer concerns, the key issue is the successful integration of products at a base level into a common suite, whilst retaining the familiar look and feel, more importantly, the benefits and values that each tool brings. Care has to be taken to consider the real cost of change for the customer. Having significant overlap creates a surprisingly difficult problem, especially with development teams in different continents speaking different languages.

Though Aegis, with DiPlan now, are competitors to our Valor MSS suite, I wish them well. If anyone from Aegis or DiPlan is reading this blog, let me say “please take care”, to try to limit the confusion as much as possible. Our respective customers do not need more uncertainty at a time when many of them are competing for survival. Bringing value is much more important than ticking boxes. Value and stability.

Tags: , , ,

A fresh look at the latest manufacturing systems technologies set against a backdrop of common-sense Industrial Engineering know-how. Michael takes us through the existing opportunities that are here today and coming up soon.