Semantic BPM part three

In part one, I looked at the limitations of current enterprise systems to build a social process ecosystem, in order to  architects, managers and company employees can understand what is the business environment, what means the concepts used within the organization and what is the business domain (context) all about.

In part two I introduced the importance of how unstructured work increases the importance of semantics, and defined a model for Semantic BPM.

To create a domain ontology is necessary to apply a set of techniques used in social network analysis. These techniques are based on the same mathematical principles of algebra as described in [15]. After all informational entities are identified it can be presented through affiliations with business processes using an incidence matrix.

In this example, let’s consider an accident management process executed inside of an insurance company.

 

Customer

Policy

Contract

Accident

Invoice

P1

1

1

1

0

0

P2

1

1

1

1

0

P3

1

0

0

0

0

P4

1

1

1

0

1

This matrix is intended to identify the relationship between processes and informational entities, is not intended to specify the type of operation that is performed on top of the informational entity such as in a CRUD matrix.

In this case we can see that the process P1 only handles informational entities related to Customer, and Insurance Policy, while the P2 process handles all entities Process P1, plus Accident.

It is important dot not confuse the concept of Informational Entity: any person, place, concept, thing, or event in the context of the business, about which information is necessary to keep the informational attributes of the entity itself. For example, the informational entity Customer has the attributes: “Number”, “Name”, etc..

One of the conclusions is that matrix representation can easily identify which business processes manipulates entities, making it easier to create a meaning.

From the above matrix it can be derived two adjacency matrices:

One that relates the intensity of the connection between the processes and the shared informational entities:

 

P1

P2

P3

P5

P1

3

3

1

P2

3

2

2

P3

3

2

1

P4

1

2

1

This matrix identifies the intensity of the data exchange between processes through access made to the informational entities, in other words shows the number of informational entities that are related to particular process and can be represented in the following graph:

In this graphical representation it can be concluded that the processes with more relationship intensity is between the P1 and P2 and P1 and P3. In the first case both processes share informational entities Customer, and Insurance Policy. The same method can be used if instead of a Process you have a Case (for Ad-Hoc processes).

Other matrix relates the intensity of informational entities together.

 

Customer

Policy

Contract

Accident

Invoice

Customer

2

2

1

1

Policy

2

3

2

1

Contract

2

3

2

2

Accident

1

2

2

0

Invoice

1

1

2

0

This matrix identifies whether a relationship exists between the informational entities and how many processes are used simultaneously. In this example the entities Insurance Policy and Contract have a strong relationship because they are both found in three processes.

The reasoning we have followed allows us to make the same kind of analysis with Activities and Tasks. What may be seem like a mere mathematical exercise is very important as we intend to involve hundreds or thousands of people to work effectively, self-directed, and socially motivated. Although structured processes theoretically force participants always to perform the same steps with the same sequence we know that these artifacts do not always have the same meaning for people. With the emergence of tools that allow Ad-hoc process execution is extremely important for people that access/manipulate certain artifact being sure of it’s meaning, otherwise we are opening the door to knowledge chaos because people with his human nature sometimes like to designate differently the same concept. This finding also opens a way for the need to define a data repository model, the ontological foundation, that is outside of the scope of these articles series, still this reference can be usefull [24].

Returning again to the hiring process flowchart example, let’s consider what we have here two hiring Ad-hoc processes excuted at different times.

 

Acknowledge application

Analise application

Examine Process

Evaluate

application

Send

application

Apply for job

P1

P2

1

0

1

0

0

1

1

1

1

0

0

1

It is clear the importance of getting the information structured this way. Two instances present differently the same concept: Application / Process. What is the dominant concept in the organization? What name should be adopted? If we have to run a report how to do it?

Using the same principle, we can analyze the dependencies between process participants. This aspect is important because through a analysis of this type we can:

  • Understand dependencies and the dependence size of the participants, ie the effects of human socialization itself;
  • If the process is executed between the ones that should involved in execution or is being executed by the ones that don’t have knowledge to do it;
  • Have a clear idea if IT supports the data flow, if there are too many hand-offs, if people depend heavily of a particular individual, if the transactions are balanced or some leveling is necessary, people are qualified to perform the job;
  • What are the knowledge sharing patterns within the organization.

All this can be achieved without having to make analysis to organizational charts and process swimlanes. Stabell and Fjeldstad [17] define a clear focus on the need to answer the questions, one of the most interesting references about how work can be executed under the knowledge management umbrella. If we look at the intrinsic nature of knowledge work there is a considerable amount of intangible communication flows such as: advice, information sharing, which are activities rarely found in flowcharts.

Thus we can understand if we look from a social perspective the roles that interact in the process so you may have an idea of organizational network analysis. This analysis may include roles performed by partners, suppliers and customers, truly end-to-end.

Returning to our initial example, considering a matrix that relates processes of an insurance company and the participants involved in these processes.

 

Contact Center Front office

Customer Account Manager

Contact Center Manager

Accident Manager

Contact Center Back Office

 

P1

1

1

1

0

1

P2

1

1

0

0

1

P3

0

1

0

1

1

P4

0

0

0

0

1

If we make an analysis using a intra dependency matrix between participants we get:

 

Contact Center Front office

Customer Account Manager

Contact Center Manager

Accident Manager

Contact Center Back Office

 

Contact Center Front office

2

1

0

2

Customer Account Manager

2

1

1

2

Contact Center Manager

1

1

0

1

Accident Manager

0

1

0

1

Contact Center Back Office

2

2

0

1

Using a graph, we get:

The graph provides these insights:

  • Back Office is the participant who supports most of the information exchange;
  • Back Office, Customer Account Manager and Front Office rely heavily on each other;
  • There is not much interaction between the Accident Manager and the remaining participants (should be?).

From a global viewpoint and applying techniques of social network analysis it’s possible to identify:

  • Do we have “stars” in our workforce?;
  • If there is a strong coupling between the connections (and if this coupling) is aligned to the nature of information exchange [18]?
  • Who plays a particular role? Is this according what the enterprise thinks it’s best?
  • What are the communication channels between individuals involved and who is more dependent on whom?
  • Do we have balanced connections?

The answer to these questions leads us to conclude much about the nature of the alignment of human resources and business processes.

In part four I will present a wrap up and final thoughts.

References:

[15] Theory and Method of Social Research – J Galtung, Columbia University Press, ISBN – 0-231030-88-6

[17] Configuring Value for Competitive Advantage: on Chains, Shops, and Networks – Stabell, C. and O. Fjeldstad (1998). Strategic Management Journal 19

[18] An Ontology Framework for Semantic Business Process Management – Martin Hepp, Dumitru Roman

[24] Ontology reasoning with large data repositories – Stijn Heymans, Li Ma, Darko Anicic, Zhilei Ma, Nathalie Steinmetz, Yue Pan, Jing Mei, Achille Fokoue, Aditya Kalyanpur, Aaron Kershenbaum, Edith Schonberg, Kavitha Srinivas, Cristina Feier, Graham Hench, Branimir Wetzstein, Uwe Keller – Ontology Management – Springer – 978-0-387-69899-1

Advertisements

5 thoughts on “Semantic BPM part three

  1. Pingback: Semantic BPM – Part One « End to End BPM

  2. Pingback: Semantic BPM part two « End to End BPM

  3. Pingback: Semantic BPM – epilogue « End to End BPM

  4. Pingback: New Killer Star « End to End BPM

  5. Pingback: Enterprise Architecture Handbook part 6 | End to End BPM

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s