Match processes to business needs: Apply BELA to case management

Tien Nguyen, Bethesda, Maryland
Executive IT Architect - IBM Software Services Federal, IBM Software Group

Marc Fiammante, La Gaude, France
Distinguished Engineer and Chief Architect, Office of the IBM Software Group CTO, Europe

Case management evolved from the necessity to evaluate, facilitate and execute on critical decisions while utilizing information in disparate repositories, departments and agencies and abiding by mission-critical guidelines and policies. These decisions must be made swiftly and accurately. Case management solutions can be deployed not only in traditional legal or social services environments but can also be deployed cross-industry. These solutions also apply, for example, to insurance claims, mortgage loan applications, customer complaints, incident reporting, and so on.

Case management as an IT solution is new. Most case management systems are custom database-centric applications addressing a specific vertical niche such as social services, legal litigations, medical care, and so on. Recently with the maturity of the technologies supporting case management capabilities, such as:

• Enterprise content management (ECM),
• Business process management (BPM),
• Business analytics, and
• Service oriented architecture (SOA),

there is a burgeoning trend toward considering case management as a generic, cross-industry IT solution.

Due to the predominant role of business processes in case management solutions, IBM Service-Oriented Modeling and Architecture (SOMA) has been a good candidate methodology because it focuses more on integrated capabilities than on pure product integration by enabling the identification of those services that will support business functions and use cases. One important activity pattern in SOMA is Business Entity Lifecycle Analysis (BELA). BELA allows for the identification of services/functions based on the analysis of key business entities’ lifecycles to ensure that the solution implementation tracks closely to the business needs.

This article describes how BELA applies to case management solutions.

In an IBM developersWorks article on Data4BPM^*, Prabir Nandi et al introduce the concept of business entities as a key representation of the business view of data and information.

The authors state that, “in most business process management tool suites data is treated mostly as an afterthought. Activities and their flows are the main abstractions and the data manipulated by the processes is essentially hidden in process variables. The presentation and aggregation of data is handled outside of the process definition, and implemented through generic service calls. This process-only approach ignores the important data perspective during business operation analysis, often obscures key aspects of the operations, and can lead to costly re-factoring throughout the solution lifecycle. ”

The authors advocate including data entities in business process modeling and introduce the concept of business entities. Business entities are defined as:

Business Entity Lifecycle Analysis (BELA) is a capability pattern built into IBM Service-Oriented Method and Architecture (SOMA) to track the state transitions of business entities against the lifecycle of the business process being modeled. BELA helps inject the business essence into process modeling by analyzing the lifecycle of business entities and determining the transition of process activities based on entity states. The various state changes in business entities help identify the services needed to drive the process to completion.

Figure 1. Role of BELA in process modeling (For full-size view, click on image)

In case management, the scope expands beyond the process dimension. Case management is a solution framework that brings people, information, and process together in order to execute specific business-mission-related projects. In that sense, BELA plays a role in each of those areas. BELA helps in the information space with accurate data models, in the people space with services and task definitions and in the process space with identification of milestones and key performance indicators (KPIs).

Figure 2. Role of BELA in case management (For full-size view, click on image)

As of now, there is no specific prescriptive approach to case management solutions development. Currently the implementation is treated as any other custom application development and uses a generic methodology. Often times, the solution is leveraging an existing legacy enterprise system, and the approach is tainted by that specific system’s biases and methodology. For example:

As a result of the lack of methodology in the domain, the tools end up defining the solution instead of the solution defining the selection of the tools. The case management problem is redefined to retrofit into the existing capabilities of legacy systems and infrastructure. This canned approach to design and solutioning does not lend itself to thorough problem decomposition and puts too much emphasis on implementation versus actually solving the business problem.

The lack of methodology often leads to a misalignment between the problem and the solution. Here are some typical activities an implementation team might find themselves engaging in:

• They all start with a swim lane diagram… focusing on the WHO and the HOW instead of the WHAT….
• They tactically dashed to the end-result instead of strategically analyzing how the transformation in the business information during the process could help reveal the real problem. For example:
  • Resolving a caseload problem: They focus on quantity and incoming load instead of the average duration of a case lifecycle, which also consumes time/resources.
  • Streamlining a manual process: They focus on digitizing content and communication and removing the “manual” part of the process. They should try to reengineer the process around the business information collected, instead of the human tasks at hand.
  • Enforcing the integrity of a process: They focus on managing the mandated part of the process such as deadlines and milestones, instead of analyzing the interim states that lead to that milestone or affect that deadline.
  • Tracking and monitoring: They focus on open/closed status, number of cases processed and overall process duration, instead of examining the dynamic changes in the business information and its quality.

The rationale for applying BELA to the implementation methodology of case management solutions lies with the prospect that BELA would:

• Provide a methodical approach for assessing the “as-is” business process and engineering the “to-be” business process.
• Provide insight into why things should or should not continue to be done the way they are done today.
• Provide a focus on the deliverables rather than on the process.
• Provide a balance in the process-centric nature of the business process reengineering.
• Provide an iterative feedback loop for validating the process and ensuring that it supports the business mission.
• Provide a means to identify patterns and inconsistencies in the process (e.g., why is there a decision point at this juncture?).
• Provide an understanding of the natural progression of business information that will help teams build rules to automate the process and reduce the human tasks at decision points.

At its core, BELA in case management is implementation-centric and focuses on services. As shown in Figure 3, business entities’ attributes contribute to the solution information model, and their states help validate the steps in the process model. The access control criteria to information are modified during BELA to maintain a business entity’s lifecycle integrity, and the conditions for state transition within an entity’s lifecycle can feed into business rules that govern the solution process.

Additionally, BELA in case management is also management-centric and focuses on feedback and control. Figure 3 shows the extended capabilities in case management that are created, validated and optimized using the BELA prescribed activities.

Figure 3. BELA’s role within case management solutions (For full-size view, click on image)

The following aspects in the implementation of case management solutions are enriched with the BELA activity.

External data transform maps

• Most business entities for case management come into the enterprise from external sources instead of through manual data entry.
• Understanding the intended business entity lifecycle as it is owned by the enterprise helps the capture and transformation of relevant attributes from external data feeds.

Content management

• One of the objectives of case management solutions is to streamline the paperwork and manage documents and digital artifacts in a central location so as to facilitate eventual discovery processes.
• Some of the business entities are not only represented as an entry in the database with field attributes, but are also embodied in one or more documents.
• The lifecycle of the business entity should be mindful of the parallel lifecycle of the document that is its representation.

Compliance/audit

• Most case management solutions are created in response to legal and policy mandates.
• Some business entities’ state transitions in the lifecycle are to be logged and reported. Other business entities’ lifecycles are to be extended and kept as enterprise records for a mandated duration.

Reporting/analysis

• Case management is about management. Depending on the reporting/analysis requirements, business entities may carry additional attributes and go through additional states transitions.
• On the other hand, understanding the business entities’ intended lifecycles will help management fine-tune their reporting requirements.
• Longitudinal analysis also requires a longer business entity lifecycle.

UI model

• To support human tasks in case management, some of which determine the state transitions of a business entity, a user interface (UI) model that intuitively represents the business entity and its attributes is crucial.
• A business entity’s attributes, its lifecycle, and its access control policy dictate how the UI will look, behave and how the field validation will occur for data entry tasks.

Collaboration

• When a process has “Split” and “Join” paths, it signals a collaborative task involving many parties simultaneously. Typically business entities are being created either in a hierarchical subtask mode or in different functional domains.
• BELA helps in determining the state interdependence across multiple business entities that is created in a collaborative mode, and drives the rules of collaboration, especially in a multi-author environment.

Human tasks/human decisions

• If a process is automated from end-to-end, it would not be called case management. Human tasks, and especially human decisions, drive the process in case management.
• Most business entities go through state transition based on human decisions. BELA helps identify the point where a human decision is needed, what those decision types are, and the process path for each.

Alert/monitoring

• Real-time feedback on the process, bottleneck corrections, and overdue task prevention are part of a case management solution’s objectives.
• Key state transitions from the business entity have to be identified and flagged for monitoring. BELA helps identify the critical business entity states and their impact on the overall business.

Triggers and events

• The convergence of various interdependent business entities in their state transitions generates events that will trigger the behavior of other business entities and the path direction of the overall process.
• BELA helps identify those asynchronous trigger points and the ensuing business rules.

In the example in Figure 4, the business entity is the grant application itself. The grant application transitions from one state to another based on the progression of the grant review process and decisions made within that process. The grant application business entity also comprises document business entities such as budget proposal, business case narrative, past performances, and so on. Those entities also have their own lifecycles depending on the overall process and the lifecycle of their parent business entity (grant application).

Figure 4. BELA diagram for a Grant Application Processing example (For full-size view, click on image)

Case management solutions help streamline business processes and human decision-making based on specific information sets and/or content items, i.e., business entities. These entities will ultimately form the basis of all enterprise mission-critical records. In the end, these business entities and the information they carry reflect the quality of the overall business process, because they are invariably embodied as both the triggers and the outcomes of the process itself. All management capabilities such as auditing, reporting, analytics and monitoring, and all business rules and events rely on the information within the business entities. This very fact highlights the importance of juxtaposing the BELA activity with the process modeling activity in order to ensure the highest fidelity of the process to the business itself.

Additional information:

Richard Hull, IBM T.J. Watson Research Center, Enabling Smarter BPM through Business Entities* with Lifecycles (PDF, 889KB), Keynote Talk presented at CASCON 2010.

J. K. Strosnider, P. Nandi, S. Kumaran, S. Ghosh, A. Arsanjani, Model-driven synthesis of SOA solutions, IBM Systems Journal, Volume 47, Number 3, Page 415 (2008)

Craig Le Clair and Connie Moore, Forrester Research, Dynamic Case Management — An Old Idea Catches New Fire, (link resides outside of ibm.com) December 28, 2009.

David Yockelson, Advanced case management and BPM: Better together, IBM Smart SOA Newsletter, June 23, 2010.

Ann All, IT Business Edge, Case Management Is Step Forward in BPM Evolution, (link resides outside of ibm.com) March 8, 2010.

*Prabir Nandi, Dieter König, Simon Moser, Richard Hull, Vlad Klicnik, Shane Claussen, Matthias Kloppmann and John Vergo, Data4BPM, Part 1: Introducing Business Entities and the Business Entity Definition Language (BEDL). A first class representation and specification of data for BPM applications, (April 2010).