Case Management Modeling: Creating Systems of Engagement
Process modeling. Most case management systems, however, employ processes. They may or may not be end-to-end processes, but there will probably be some. Here we employ the concepts of "state modeling" and link the states together to form processes. We can make the process very rigid (like BPM) and only allow our cases to flow from state to state in a fixed order, or we can be very free-flow and give the knowledge worker the ability to choose which state a case should reside in at any given time, or (as is most common) some combination of the two.
Returning to our above example, faced with the tardy employee the manager may have several states available to him that he can move the employee's file to. One such state might be an evaluation state for disciplinary action. The employee's file could be automatically moved to the state in process fashion (triggered by their tardiness) or the movement to the evaluation state could be manually triggered by the manager-it would depend on how the ?system is designed.
Dynamic case modeling. An additional benefit of a system that begins with a data model is that it can generally evolve over time adding data objects, elements, rules and processes as needed. Users may find that after using the system for a time that there is an essential piece of data that they have not been capturing. In a dynamic environment this element can be added to the data model and will become part of the overall system design. Dynamic case management systems also facilitate the incremental addition of rules and processes. Returning one last time to our above example, if we find that tardiness is becoming an issue and that managers aren't addressing it consistently, we could establish a rule that automatically moves an employee into the evaluation state after a fixed number of late arrivals. Now we have a fixed process where one did not previously exist.
Why BPM is Challenged by Case Management
BPM platforms begin with process modeling and move up the stack, adding policies and finally data—a fine approach if you are building a rigid, process-driven system. If there is going to be a lot of unstructured workflow guided by knowledge workers in the case management system, starting with process modeling is going to be challenging. Additionally, end-users don't think in terms of process so engaging them with a process-first approach will always be a challenge.
At their core, case management systems are data intense. They manage the investigations, appeals, claims, correspondence and other matters that affect our daily lives. If the data isn't correct or comprehensive and the relationships aren't complete, it doesn't matter how good the process is... the system won't help generate appropriate outcomes for the constituents it serves. Fortunately end users can identify the information that they require to accomplish their work. Beginning with a data-first approach facilitates a dialog between system designers and end users that ultimately leads to enhanced systems of engagement.
Case Modeling Starts with a Data-First "Information Model"
- Case modeling and development begins with data/information (structured and unstructured) organized around a folder (a case object);
- The folder is the case. It is able to contain all relevant case information;
- A case progresses through "states" (aka: milestones, goals);
- Process modeling comes later—if at all; and
- Cases evolve dynamically as new data, events, and tasks ?unfold-and map and maintain relationships with these data objects.
MicroPact's entellitrak is a unified BPM platform that takes a data-first approach to dynamic case management and business process management and is available with fully integrated document management, natural language analytics and mobile modules.