At the Gartner Portal and Collaboration conference in London I had the opportunity for a long conversation with Gartner Fellow Tom Austin about Pattern-Based Strategy and People-Centric Computing. I was surprised that he was surprised that I felt that these two subjects are not only related but identical. For me, People-Centric Computing is just another word for managing human interaction in business processes.
Gartner sees ‘Pattern-based strategy’ implemented in Intelligent Business Operations: “Intelligent business operations are a real-time way of applying Pattern-Based Strategy in the context of transactional and record-keeping business processes.” Gartner acknowledges here that identifying patterns of business events tied to business processes is viable knowledge. They do however not consider the use of pattern matching technology but propose that the patterns are manually identified and encoded in rules and flowcharts that trigger from data and complex events. Thus, CEP Complex Event Processing and CBE Complex Business Events are essential elements of Gartner’s intelligent Business Operations perspective.
Gartner: “Intelligent business operations are a style of work in which real-time analytic and decision management technologies are integrated into the transaction-executing and bookkeeping operational activities that run the business. The relevant technologies include business activity monitoring (BAM), event-processing, interactive and graphical business dashboards, report writers, decision management with rule engines, constraint-based optimization are used in real-time mode in operational decision making.”
The Difference between BPM and IBO
I don’t see BPM and IBO as ‘mutually reinforcing’ but that IBO is clearly NOT referring to high-volume, low-value process automation with BPM, but describes high-value, low-volume decision making and innovation processes performed by knowledge workers. Knowledge processes are clearly people-centric and require empowerment through goals, authority and means (of which the IBO platform is one). This means that an IBO environment requires ACM-like capabilities for supporting human interaction and that should involve goal orientation. As it happens, I haven’t seen any mention of goal-orientation in IBO yet. The performers are in the drivers seat, not the agents, flowcharts, rule engines or predictive analytics. But yes, I agree that the complex events, decision making and human interaction need to be tied together.
Nature gave humans intuitive capability for decision making, meaning to apply practical plausibility to repeatedly observed patterns in complex adaptive systems without being able to ascertain their true causality. Human decision making can’t be predicted because the actual mindset is intangible. BPM and rule automated decisions are an expensive management mistake where they are meant to manage or control intangibles (Peter Kruse). Therefore, I consider predefined flowcharts, automated decisions and predictive analytics as unsuitable to optimize real-world human interaction in knowledge work. I propose that human decisions can and must NOT be automated, because too much of it depends on assumption of models, measurements and algorithms. Intelligent Business Operations must be meant to SUPPORT human intuition and not substitute it.
How To Provide Strong Data Integration
The million dollar question for usable IBO is how to achieve the strong business data focus that is beyond the usual BPM ability. Each of the IBO components has to deal with business data in REAL-TIME and the mapping and passing of data from interfaces to analytics, to event recognition, to business rules, and to process management and user presentation is the real challenge. plus you need a platform that allows scalable and distributed agents to handle all of it. The solution to this problem is to provide a business language ontology that is used to express a Business Architecture with customer outcomes, strategic objectives, operational targets process goals as well as all business entities used in processes and decisions. That is only considered by Gartner for the analytics part.
Business rules and predictive analytics have to be ‘integrated’ in some unspecified way. I wish everyone much luck with that. A holistically consolidated platform is required for Intelligent Business Operations. Let’s not forget business content that is the key element in business events and taking action. And how will the analytics and complex events link to the data of running processes in real-time and identify complex business event patterns?
Well, that has been the core of my research over the past ten years and resulted in 2007 in the patented User-Trained Agent that learns from performers what events are linked to which process context and require which action. Data, content, processes, human actions, rules and analytic results are all kept in the same state space that the agent uses to perform pattern matching. Complex business events can be identified via SOA data feeds but they are useless if it won’t enable knowledge workers to react with freely chosen actions that are themselves input to follow-on activities. Performers and process execution must be guided by defined process goals to ensure customer outcomes. If new process templates can be extracted from previously successful executions, then the process management becomes adaptive.
The future of Intelligent Business Operations is already here. The same is true for basic goal-oriented processes. More on that in my next post.