Situational awareness is a complex concept that cannot be accomplished by any of the “CEP” software on the market today. In his post, Human CEP: the gut feeling?, Paul Vincent of TIBCO says,
[A] firefighter avoiding death by analyzing his situation …. is provided by event-driven CEP systems.
The fact of the matter is that the current “CEP” systems on the market today are not sophisticated enough to provide situational awareness. Situational awareness, generally speaking, requires reasoning and none of the software “engines” marketed as “CEP” today are capable of the sophisticated reasoning required. Let’s review Wikipedia on situational awareness.
Situation awareness, or SA, is the perception of environmental elements within a volume of time and space, the comprehension of their meaning, and the projection of their status in the near future. It is also a field of study concerned with perception of the environment critical to decision-makers in complex, dynamic areas from aviation, air traffic control, power plant operations, military command and control — to more ordinary but nevertheless complex tasks such as driving an automobile or motorcycle.
It is almost insulting to the folks who have been in the field developing reasoning systems for decades to have software companies with little to no experience in reasoning software now representing their software as reasoning expert systems using simple rule-based approaches to complex problems.
Adding to the hype in Still a question? Business rules vs CEP… we read where CEP is also central to SOA, business rules and business rules management systems (BRMS).
We read how CEP can save the world, mitigate financial disaster, save the lives of firefighters, perceive, reason and analyze situations like the human mind. We even read how this can be done with modified RETE software and extended SQL queries over sliding time windows. So let’s ask the vendors, “What can’t “CEP” engines do?”
Paul Vincent says:
Wednesday, February 25, 2009 at 10:51am
Hi Tim – the following may be useful in this discussion: “CEP will help advance the state-of-the-art in end-to-end visibility for operational situational awareness in many business scenarios” – see http://www.thecepblog.com/2007/05/14/what-is-complex-event-processing-part-1/
Cheers
Tim Bass says:
Wednesday, February 25, 2009 at 12:24pm
Hi Paul,
Yes, you are right; but the “CEP companies” are not engaged in this dialog, unfortunately. Hence, we have the right worldview, as does TIBCO, but the progress is painfully slow.
BTW, I have it on my “to do blog list” to write a post about TIBCO’s progress in merging Spotfire and Splus from the Insightful acquisition. I read some very encouraging things on the TIBCO web site recently about this (Spotfire and Splus) and have not had time to do my homework and write about it.
Thanks for stopping by and commenting.
Yours faithfully, Tim
Tim Bass says:
Wednesday, February 25, 2009 at 12:29pm
Note: Thank you Richard Veryard, Everware-CBDI International, London, for pointing out a typo I made (and have corrected):
“not sophisticated enough to not provide situational awareness”
should have been (sorry, I typed a double negative!)
“not sophisticated enough to provide situational awareness”
Thank you Richard!
Your sincerely, Tim
Roland de Boo says:
Wednesday, February 25, 2009 at 12:33pm
Hi Tim,
I’m currently writing a thesis on CEP. I was just reading about the JDL data sensor fusion model (a blackboard architecture) when I realized a nice analogy which I thought I’d share.
This morning an airplane crashed in the Netherlands; I was tracking the event by refreshing the associated wikipedia pages.
It demonstrates (I think) all kinds of aspects of creating situational awareness: independent agents provide their own observations, which are often contradictory: for example, Turkish officials report no casualties, while Dutch television speaks of 9. Some 3 hours after the event this basic factual information has not been resolved.
Meanwhile other agents are annotating the wikipedia pages with relevant background information about the airplane, the company, the crash aftermath, similar events, and speculations about the causes of the crash.
It’s fascinating because you can literally see the situational awareness growing. At the same time I cannot conceive how this process could be automated – certainly not with the current technology.
Cheers
Tim Bass says:
Wednesday, February 25, 2009 at 2:27pm
Hi Roland,
Thanks for stopping by and visiting (and commenting).
Precisely. Uncertainty is at the heart of complexity, It is the intricate relationships, different observations, relative perspectives, errors in measurement, errors in the model, and so forth (sorry, I am working a network problem and not giving a good reply) that make true situational knowledge quite challenging for humans and infinitely more for automated processing.
This is precisely why we must solve this problem sooner than later because the network, i.e. the event cloud, grows by great numbers every day, but our ability to process and make sense from the “event soup” is very limited.
This is not only a technical problem, but also a social and political one.
Yours sincerely, Tim
Peter Lin says:
Wednesday, February 25, 2009 at 4:23pm
Here’s my bias perspective as an user and developer of expert system shell. A business rule engine, expert system shell or CEP engine at best provide a foundation for creating an expert system, but they’re not expert systems. Also, not all CEP engines are created equal. Many don’t provide inference capabilities.
To create an expert system, one has to create a knowledge base, which includes data model, facts and rules. Even then, for dynamic environments, the classic knowledge base system approach isn’t enough. A knowledge base system by itself, it’s not well suited to highly dynamic environments.
As we’ve discussed on this blog in the past, machine learning is needed to handle dynamic environments. Not all rule engines or CEP engines are capable of machine learning. This is true of any engine that compiles rules down to Java code or C# code. Many, if not most machine learning systems choose an interpreted design for this reason. Whether a CEP is capable of machine learning really depends on it’s design and implementation.
My bias perspective, proper support for dynamic SA would require machine learning and statistical analysis capabilities. Without that, it’s just business decision automation. I’m sure others will disagree.
peter
Tim Bass says:
Wednesday, February 25, 2009 at 7:23pm
Hi Peter,
I certainly agree with you.
Thanks for stopping by and visiting.
Yours faithfully, Tim