Last month, I discussed the benefits of using simulation tools for modeling and analyzing business processes. In this Advisor, I want to discuss what I consider one of the most important developments in business process management: the ability to monitor the efficiency of distributed business processes in near real time.
Innovative vendors like Hewlett-Packard and IDS Scheer now offer products designed to monitor business processes as they execute, and then display their findings — based on calculated key performance indicators (KPIs) detailing cycle time, adherence to delivery date, process costs, and so on — in management dashboards that highlight the performance of core business processes. These applications also provide capabilities for triggering and sending alerts — in the form of e-mail and pager messages — to process owners and other interested parties. As a result, users can proactively respond to process deviations before they impact operations.
Business process monitoring applications also generate a logbook of business activity, which is stored in a process warehouse database. This information is extremely valuable to both IT and business analysts. By analyzing this information, analysts can uncover patterns and trends indicative of problems and inefficiencies in process executions. This can assist them in taking steps necessary to improve process execution quality — both as perceived by the users in terms of better and faster processes, and as perceived by service providers in terms of lowering operating costs. In addition, this information can be further analyzed by applying “process mining” analytics — algorithms that can automatically predict process execution quality problems prior to their actual occurrence. Again, this allows users to take steps to avoid problems in advance.
Suitable areas for applying business process monitoring applications include the following:
•Order processing
•Monitoring service level agreements (SLAs)
•Procurement
•Incident and problem management
•Transport and logistics
Statistical data generated by business process management systems (and the various applications they work with) can be highly complex and technical in nature. Consequently, it can be difficult for business users to interpret the data. To get around this, business process monitoring applications provide a set of easy-to-use functions (accessible via a dashboard) that allow users to define metrics that can identify process instances with characteristics of interest to the analyst — for example, because they correspond to especially low-quality or high-quality executions. (An example of a low-quality execution might be a process instance that takes more than three days to complete using a particular resource or service.) In addition, analysts can define taxonomies, which instruct the application to classify process instances according to the user. For example, users might define the classes “successful” or “unsuccessful” for a process, and specify success criteria. The business process monitoring application will then classify process instances according to these criteria and display the results to users.
The ability to analyze process instances based on user-defined classification criteria is a very powerful tool for understanding and improving process execution quality. This is because it allows business analysts — as opposed to only IT personnel — to extract knowledge about the circumstances in which high-quality or low- quality (for example) executions occurred in the past, and to use the information to explain why they occurred as well as to predict potential problems in running processes in the future. As an example, consider order processing. Users might determine the number of “unsuccessful” process executions that took place during a particular weekend or holiday — compared to their execution success rate during the regular work week, and to identify causes for the degradation. Or, they might use it to identify specific processes that did not meet SLAs as stipulated by contract and to take steps to ensure that they execute successfully in the future.
Conclusion
Business process monitoring offers organizations a number of benefits when it comes to their business process management and optimization efforts. For one, it allows companies to monitor and analyze distributed business processes as they execute, enabling proactive response to process deviations before they can impact operations. It also allows users to apply automated analyses in order to predict process execution quality problems prior to their actual occurrence. Finally, and perhaps most importantly, it can put these capabilities into the hands of business analysts — as opposed to just IT personnel (as is typically the case).
As always, your comments and insights on this announcement and enterprise architecture issues in general are always welcome. Send your comments to chall@cutter.com or call me at +1 510 848 7417.
–– Curt Hall, Senior Consultant, Cutter Consortium
