So, you have an ERP system

Many companies today have accomplished the big step of a successful ERP installation, standardizing processes, systems and hopefully leading to a better managed supply chain. But todays world is ever-changing-products come and go at an ever faster rate, promotions change weekly, opportunities are short-lived, your customers frequently re-tender their business.


How is a company to cope with this complexity? Todays ERP systems dont allow one to ask interesting what if? questions about your own business future.


What if I take on this large block of business next summer in this factory? What if I reallocate some of my production to a closer plant? What if I install a new packing line here? Which machine should I target for efficiency improvements? What will happen to my customer service levels, my warehouse costs, my transport costs, my production costs, my overtime shifts?


The solution to all of these problems is a realistic agent-based simulation tool, driven with real historical data from your ERP system, which can be used to provide accurate decision support for such strategic and tactical questions. With todays computer power on an ordinary laptop such tools can now perform extremely realistic simulations of years of operation of a broad supply chain in a matter of minutes.


The reality applying virtuality

Companies such as P&G, Unilever, SCA, Air Liquide are using such tools. A typical anecdote comes from SCAs recent use of such tools in their packaging business. In the words of the Edinburgh factory manager, Richard Sharp: We used the virtual factory to model various scenarios and this process resulted in the identification of strategy, that whilst it involved making several changes in the business, should meet the challenge in a cost effective manner and without detriment to our customer service levels. We are most pleased to report at this stage that reality has indeed turned out every bit as good as the model suggested. The plant just now came through its busiest month ever having achieved record levels of production. It produced 10 per cent more volume than the same month last year with production costs that were 13 per cent lower. They also achieved a staggering 30 per cent reduction in distribution and storage costs whilst continuing to deliver high standards of customer service.


This agent-based modelling approach is one that has grown up over the last 20 years. From its early days as a part of the academic field of complexity science (studying bird flocks, ant colonies and stock markets), it has found a surprisingly wide range of business applications in fields ranging between operational risk, crowd control, logistics/routing, manufacturing and the supply chain. The key to a successful, valuable solution in this area is the merging of serious scientific brain-power with industry expertise and understanding to create accurate, realistic, easy to use what-if tools that can pull up to date data from any ERP system and investigate the future in a totally risk-free fashion. From a slow start in the early 90s, such tools are now beginning to hit the main-stream. In many ways such tools are allowing companies which have gone through the pain of a large ERP installation to really capitalize on the benefits, deriving real value from the huge datastore they now have available.


Vince Darley is President and CEO of EuroBios UK. He started his academic life with a mathematics degree from Cambridge University, then moved through theoretical physics and computer science degrees before being seduced into complexity by the Santa Fe Institute. His doctoral dissertation was concerned with understanding the natural dynamics of large systems of autonomous, optimizing agents, and how to design local interactions in order to bring about particular global goals. He holds a BA, Mathematics, Cambridge University; a MS, Computer Science, Harvard University and a Ph.D., Complex Systems (Applied Mathematics and Economics), Harvard University.

Comments (0)

Add a Comment

This thread has been closed from taking new comments.

Editorial: +44 (0)1892 536363
Publisher: +44 (0)208 440 0372
Subscribe FREE to the weekly E-newsletter