I attended a very interesting presentation given by Henk from Cordys this week on Case Management. With Henk’s permission I have posted his slides on Slideshare.net (you can see them embedded below). Case management is an interesting “subset” of Business Process Management. I say “subset” as I don’t think that case management “processes” are really the same as the kind of processes that BPM targets.
A healthcare-centric definition from the Case Management Society of America goes something like this:
A collaborative process of assessment, planning, facilitation and advocacy for options and services to meet an individual’s health needs through communication and available resources to promote quality cost-effective outcomes.
If we adapt this to remove the healthcare focus and map it somewhat to how we tend to think about processes, we get something like this:
A collaborative process of assessment, planning, facilitation and action for activities and decisions to meet a specific organizational need through communication and available resources to promote quality cost-effective outcomes.
I highlighted what I consider to be the key words in this definition:
Because these processes often involve more than one employee as well as a customer in a series of non-linear interactions
- Activities and Decisions
Because at the end of the day activities or tasks have to be performed and decisions made if the situation that caused the creation of a case is to be resolved
- Specific organizational need
Because there is always some particular and defined reason for the case to be managed, even though the process might be loosely defined and flexible.
Because one must stay focused on outcomes in any process, even an unstructured or semi-structured one
Henk made a similar case in this presentation and went on to discuss the challenges of a standard like BPMN when it comes to supporting semi-structured and case management processes. For instance, a case management process is not really flow based and so flow modeling is very unhelpful. Like Henk I believe that case management is different because it is less structured and less repeatable than a typical process. As a result I think it is helpful to consider these kinds of processes from a data-centric, state-centric and decision-based perspective.
In his presentation Henk proposes a model for a more state/event and rules-based approach. I found this compelling, though I would talk more explicitly about decisions. Henk proposes managing lots of case-centric events, taking place over time. These events would include follow-up and milestone events that reflect status measurable from outside the process. These events could then be linked to activities using rules to control the allowed states – to define the applicability rules of activities. Each such milestone would need up to three decisions:
- “what is expected next”
- “what is allowed next”
- “is case complete”
These decisions could be left up to the case manager or slightly restricted using a rules-based decision or completely automated using a combination of rules and potentially analytics. Indeed adaptive control techniques could be applied to see which approach worked best in different kinds of cases. The advantage of using declarative rules to manage these decisions are clear – it is easier for the case worker to see why certain things are being recommended or allowed, the declarative approach means that the completion rules can be easily assessed against a constantly updated case file, and the decisions can be managed and improved explicitly because they have been separated out and identified.
Too many companies, I think, feel that their choice is between automated straight-through processing and handing over control completely to their case workers, loosing repeatability and control in the process. Case management can and should be integrated with decision management for maximum effectiveness.
Here’s Henk’s presentation:
What’s your experience with case management?