MES and the Cloud
October 5, 2010 1 Comment
Significant discontinuities occur in technology every 5 – 10 years. The development of the personal computer, client server computing and the Internet have each had a profound impact . Each development has been accompanied by significant excitement and “hype” in the industry, which can unfortunately clutters the message relating to the underlying value and as a result creates some scepticism when a new concept is mooted, and then hyped out of proportion.
Yet, when the largest technology companies in the world demonstrate a universal acceptance of a new concept then we should sit up and take notice, because this is probably beyond mere hype. And so it is with Cloud Computing.
Rather than simply jumping on the hype bandwagon, I would like to take a few moments and address a specific aspect of cloud that should concern manufacturing companies. The question is what role cloud computing will play in the evolution of manufacturing execution systems (MES). Should CIO’s in manufacturing be doing something now, to prepare for the shift that will come? What about manufacturing systems managers, process control engineers, traditional system integrators and others?
On the one side cloud computing is already practical at the business application level . For example several examples of successful and mature hosted CRM solutions have existed for several years that demonstrate the viability of consuming business critical services from the cloud. Examples of ERP systems in the cloud also exist, and the number is growing. Right now the major application vendors are not idly waiting for someone else to develop the software services market. They are all actively building cloud based application infrastructures, exploring relationships with telecommunication and other hosting providers, and promoting SAAS based software architectures that allow on-premise or hosted applications. These architectures are based on web services, a common security model and a standard for information transfer between these environments.
So lets briefly explore specific manufacturing systems: SCM (supply chain management), MES (manufacturing execution systems) and process control . Where could these fit “in the cloud”? The answer is of course that conceptually, all of these (except direct process control) can be hosted and provided by third parties though “the cloud”. The main issue with near real time process control of course relates inter-alia to the long latency of connections to hosted servers, introducing safety and other engineering considerations that make it impractical to host anywhere but on-premise. However it would be a mistake to argue that because process control does not fit into the cloud neatly, the same applies to the rest of MES. In fact, process control systems that are architected with cloud in mind will work better with cloud based MES level applications in future. And consider the example where NASA can fully control a complex orbiting space station remotely. Complex remote process control is possible, just not economically viable perhaps.
MES include such applications such as laboratory management (LIMS), operations performance management, business process automation (such as safety incident reporting, maintenance work orders and permit to work), and others. None of these are significantly constrained by the inherent latency in cloud based connectivity. In fact, service providers that can simplify standard “commodity” business processes such as work order processing (plant maintenance), or “sample analysis and reporting” (laboratory systems); and make these applications available to customers who are happy to consume these as required through the cloud; stand to capture significant market share over the next few years.
For manufacturing companies, cloud based MES solutions allow standardisation of manufacturing sub-processes across multiple plants in many countries, a concept that will be attractive to the global giants who acquire manufacturing assets around the world and seek to leverage best practices internally within their entire organisation.
So how should manufacturing CIO’s and systems managers react? My recommendation is to do what all engineers do when confronted with a fundamental shift in technology. Get informed, quickly. Study cloud computing now because it will impact on the way you will select, deploy and integrate software applications in future. In studying the subject, get your hands dirty and experiment with the new virtualisation technologies available. Look at the underlying architectures in Microsoft’s Azure to see how windows environments will be impacted. Consider identity and security models and how these will become even more complex when critical data in future is hosted outside your firewall and direct control. Consider how Look at the service providers, are they going to be able to provide you with a reliable hosted service. Consider governance, how will you meet legislation and other requirements around your information? Look at your software vendors, which of them have a clear strategy to move selected applications into a hosted model, and how will they work with the infrastructure companies going forward.
The technology world is about to make a fundamental shift (again). Manufacturing companies will not be immune to this new wave. As the saying goes, there is no smoke without fire; there is no hype without an underlying driver – be informed and be prepared for the next generation of MES. Dare I also jump on the bandwagon and call this MES 2.0?
Gavin,
Thanks for your insights into how the cloud will effect manufacturing operations. One of the benefits to cloud computing I am promoting is the ability for the big auto manufacturers to improve their ability to communize and streamline their supply chain. Too many times in today’s environment the tier suppliers have not or could not invest in plant floor applications. For many of these tier level companies the plant IT budget is controlled at the plant managers level and they must decide whether to invest in some foreign software package or invest in a new piece of equipment to improve throughput, quality, etc. If the cloud is done right we can setup a pay as used system to help with the cost justification. However, this is not the real business benefit. The actual advantage is enabling a common platform the OEM’s can offer to the supply chain in order to improve things such as visibility, inventory, warranty cost, intelligence, customer relationships, and much more. In today’s environment where many of the tier companies continue to run their plants using excel, access, plant based applications, etc these improvements are very difficult to achieve in any sort of intelligence based way. I also believe there is a significant revenue opportunity for the OEM’s in introducing such an environment. Take for example GM. They have world class data centers all over the world, but sadly for the most part these data centers are only used to support their own internal organization and is – again for the most part – not used to provide information and intelligence for their supply partners. What if they leveraged these data centers to provide cloud based applications for their partners. What cost savings could be achieved? What revenue could GM gain from such an offering? More importantly, what opportunities does this provide the supply chain to react to market change, partner on new opportunities, and deliver more cost effective quality products?
I believe the cloud goes way beyond simply reducing IT cost by eliminating hardware and software license cost to the OEM, it provides an opportunity for a significantly improved and transformative way of doing business.