There is a strong possibility that if you are reading this article you will be reading it in a magazine. Words printed on paper are sent through the post, transported by train or in a van and then delivered directly to your mailing address, carried in a mail sack by a postman.
(Web editor’s note: This article is reproduced from the IMS supplement to ‘The Engineer’ due out 25/5/01. Magazine editors are entitled to an opinion, I guess…)
The point is, everyone is talking about e-business, and in our specific domain, e-manufacturing, but the extent to which it is being achieved remains modest. Magazines like this one have not been conceived to barrack you about what you should and should not be doing. We understand the limitations.
IMS too exists on the web but we know it will mostly be read on paper. We have yet to see a paper-free office anywhere, although people have been talking about the concept for at least a decade. So why, right now, should manufacturing be any different?
Many manufacturers are still grappling with the concept of getting started in e-business, anxious that they are way behind where they should be. ‘If you have only just installed touch-tone phones around your plant, then it is a safe bet that putting in an e-manufacturing system is likely to be a long-term plan,’ said Kevin Prouty, a former plant manager at a US car manufacturer.
Prouty is now one of the many engineers-turned-analysts advising on e-business for US manufacturers. His team of number crunchers at AMR research believe that industry is on the verge of a major spending binge on IT, infrastructure and automation that is all geared towards gathering information from the shopfloor and integrating it with the rest of the enterprise system. Big spending on enterprise resource planning systems will slow, they predict. But spending at plant level on IT and infrastructure is forecast to grow by 20% in the biggest US companies over the coming year.
It is easy to see why. ERP systems have brought efficiencies in book keeping and tracking manpower costs. But there is a perception that they fail to get under the skin of the manufacturing operations, imposing a rigidity in planning based on assumptions about plant capacity and the effectiveness and reliability of the supply chain. Hence the jibes about expensive consultants spending months examining a business and then installing an ERP system that is akin to pouring quick-setting cement over the entire business.
Because of these limitations, most companies either have not made the investment, or have under-invested, to tightly couple plant floor manufacturing directly with the ERP system. As a result, one of the biggest challenges facing the effectiveness of enterprise systems has been to couple them to the realities of the plant floor: the now commonplace notion of ‘shopfloor-to-topfloor’.
Getting investment for such developments has remained tricky up to now. A survey by Benchmark Research, published at last November’s CIM show in Birmingham, showed that the vast majority of spending on a broadly defined grouping of e-business projects is being driven by sales and marketing departments: basically website shop-windows and on-line catalogues with orders routed to existing business processes. Compared with the reduction in costs achievable through better efficiencies in purchasing and manufacturing, it’s small beer, and in fact on-line ordering quickly risks problems of unfulfilled orders if not linked to the capacity built into the manufacturing process – the bit of the business that is usually branded ‘old economy’.
But if the spending on this part of the business is about to take off as analysts like AMR and others predict, then this situation will quickly change.
One reason is intense interest at board level in the level of asset utilisation. Measures such as return on net assets are part of the mix that influence investors in deciding just how sexy are the shares in a quoted company. Another boardroom fixation is increasingly becoming the costs of maintenance and downtime.
In highly capital intensive operations, a company can find itself paying out almost as much in maintenance as it does in tax. Little wonder, then, that automation companies are starting to create a ‘boardroom sell’ that aims to get top-level support for investment plans at the plant floor.
‘In a sense, these kind of investments are really presenting themselves as a financial story with a technical back-end,’ said Andy Bates, European Business Development Manager based at Rockwell Automation’s UK office in Milton Keynes.
The game now is for major manufacturers to own as few assets as possible – hence the popularity of outsourcing, shifting manufacturing off the balance sheet, turning it into an operating cost – and to make the assets that are still there work as hard as possible.
It is here that the real shopfloor-to-topfloor integration needs to take place. There are probably dozens of different areas of technology that could boost asset utilisation, and hence return on assets. Two obvious ones are scheduling and preventative maintenance.
If suppliers really are going to take part in internet-based auctions to win new business, then they are going to have to plan their manufacturing strategy during the course on an on-line auction. This is not impossible, but it requires a continuous flow of information about asset utilisation, or manufacturing capacity. Scheduling and maintenance systems will just about do it, and are available on the market now.
The basic idea is that incoming orders are routed by the existing enterprise system to the scheduler, which works out when they can be made. A refinement on this would be to weight the importance of orders by the level of profit they will create and then add a number of constraints, such as the level of flexibility of delivery dates. It’s not rocket science, though to do this with style, then incoming orders would be best supplied in a ready-to-input format such as XML, or via a web-based ordering system on the supplier’s website.
If the scheduling system is a way of providing a sophisticated picture of incoming demand, then the maintenance system does the same for supply. This time, the investment can be more incremental, simply by replacing existing shop-floor devices with upgrades that capture a bit more information about what’s going on.
Rather than just signalling whether it is on or off, for example, a component could add further details about its own operation, giving an indication about how long it has got left before it breaks. Computers that monitor all this information can then order replacement parts, and schedule the maintenance to minimise downtime. This information is also shared with the scheduling programme and juggled to find an optimum time to do the repair.
It’s an appealing vision, but it’s also essential if the ‘open-all-hours’ web-based ordering which manufacturers all want to achieve is to be realisable.
Of course, getting from existing manufacturing to a model like this is the hard bit. While the components to make this happen do not of course all have to be installed at the same time, dealing with all the existing information and business processes around the different sites and departments within a manufacturing plant is one of the most problematic parts of the operation.
Similarly, finding a supplier for a long-term operation is time-consuming and costly. With the individual physical components being just one part of an overall solution that is more akin to business consultancy, the automation supplier’s engineers become an almost permanent fixture on the customer’s site – eating in the same canteen, drinking the same coffee, using the same sports centre, though, as some say, gratefully, attending fewer meetings. Not surprisingly, the trend now is to sell the entire automation system on the basis of its total cost of ownership.
This in itself can provide a highly complex set of competing measures and tariffs among competing suppliers. Each will be citing savings in down-time, better utilisation of installed capacity, as well as various deals on consultancy, back-up, helplines, maintenance agreements and so on. It makes the considerable hassle of choosing a mobile phone provider look like child’s play.
And the money involved in considerable. Mark Hughes, engineering consultant at US pharmaceutical giant Eli Lilly said his company spent $600,000 during 1993 working with four rival suppliers, just trying to choose an automation platform. ‘To get a large organisation to agree with such a decision, and to stop sending you e-mails about it all the time – that takes time,’ he said.
The costs and complexity of the service being offered is another major driver in the tendency among automation suppliers to consolidate. Major manufacturers are trying to limit the number of suppliers to just a handful of major partners. Some suppliers are also trying to form links with global business consultancies, adding the web-enabled manufacturing plant to part of the continuing development work on the enterprise system.
But it is still very early days. Only about one company in a hundred are making inroads into the kind of scheduling and maintenance ‘loop’ that will allow quick response to orders gathered over the internet. But that is set to change, probably within the next three to four years.
And perhaps by that time, we can expect to have rather more readers of articles like this downloading them from the web, rather than waiting for the post to arrive.