A Synthesis and an Extension

The third wave of BPM is not a fantasy, a false promise or hype. For BPM, like other true breakthroughs, is based in the mathematics, specifically process calculi, the formal method of computation that underpins dynamic mobile processes, as opposed to static relational data. Pi-calculus, one branch of process calculus, has recently drawn considerable attention in the computer science community and by those building process management systems. The underlying semantics of BPM, the business process modelling language, must be based on an open standard available to all participants (people and computer systems) in a value chain. The radical breakthrough is that in the third wave, business processes are directly and immediately executable - no software development needed!

BPM doesn't speed up applications development; it eliminates the need for it. Without its mathematical foundation, businesses would be correct in thinking that BPM is just another buzzword, acronym or marketing ploy. To make BPM a reality, its underlying business process language must be rich enough to describe the process of hosting a dinner party yet also precise enough to describe how computer system "A" talks to computer system "B" while computer system "C" may drop in or out of the conversation, in the same way participants do in real business processes.

The essence of the BPM innovation is that, based on the mathematics, we now understand data, procedure, workflow and distributed communication not as apples, oranges and cherries, but as one new business "information type" (what technologists call an "abstract data type") - the business process. The recognition of this new fundamental building block is profound, for each element in a complete business process (the inputs, the outputs, the participants, the activities and the calculations) can now be expressed in a form where every facet and feature can be understood in the context of its use, its purpose and its role in decision making. This problem-solving paradigm can therefore provide a single basis not only to express any process, but as the basis for a wide variety of process management systems and process-aware tools and services. Some of these are already available; others will be developed in the future. The implementation of such technology has required a re-examination of some deeply entrenched common wisdom, such as the notion that software is always built from objects and components. Now we can "develop with processes" as well as "manage with processes".

Going forward, this new information type and its associated management systems will be far more important than the relational data model and its associated database management system that underpins the vast majority of today's business applications. The new information services that implement this approach can read, write, query, compose, decompose, transform, measure and analyse end-to-end business processes, internally, with business partners and in the context of external information sources.

BPM enables businesspeople to manipulate familiar business processes directly and provides the ability to conduct what-if analyses to optimise results. No programming needed - simply design, and, presto, execute! BPM takes software development off the critical path of business process management, and off the critical path of business change and innovation. Do not conclude that BPM is a lightweight solution suitable only for trivial tasks. BPM encompasses a mission-critical infrastructure equal to, or exceeding, that of today's massively scalable, fault tolerant, data management and transaction processing platforms. Welcome to the next fifty years of business and IT.

Howard Smith is chief technology officer (Europe) of Computer Sciences Corporation and co-chair of the Business Process Management Initiative. Peter Fingar is an executive partner with the Greystone Group. The authors can be reached at authors@bpm3.com