“When you have a serious problem, it’s important to explore all of the things that could cause it, before you start to think about a solution.
That way you can solve the problem completely, first time round, rather than just addressing part of it and having the problem run on and on.
Cause and Effect Analysis gives you a useful way of doing this.”
The video below provides an example of using fish-bone diagrams to identify the root-cause of a problem by working back from all of the factors that could be causing the problem. In a network operations centre (NOC), staff are making these types of determinations on an everyday basis.
I wonder whether there’s an opportunity to build a learning system around this model with the following characteristics?
- The learning system is installed and integrated with an alarm management system
- Right clicking on an alarm or group of alarms would allow an operator to bring up a fishbone tool and fill out all the possible causes to establish a fishbone diagram for this event
- The learning system would be calibrated by having the customer’s experts build up fishbones for each event type over the period of a few weeks
- The learning system would get progressively better at showing fishbones for each style of event
- The operators would also use the fishbones to attempt to resolve root-cause of the events
- They would record the correct results (ie verified root-causes) in the learning system and progressively calibrate the system to give responses to all recorded event patterns
- The learning system becomes a decision support system that gets progressively better at highlighting the real root-cause, possibly even automating the process to an extent
- The learning system could be integrated with problem / ticketing management systems to kick off standardised remedial activities / processes for most recognised events
Could this work? What are the limitations?Read the Passionate About OSS Blog for more or Subscribe to the Passionate About OSS Blog by Email