Collaborative Workflow Management for Logistics

Consortium

Leo Pudhota, Elizabeth Chang, Jon Davis, & John Venable

School of Information Systems Curtin University of Technology

PO Box U1987, Perth WA 6845, Australia

Abstract. Logistics activities require strong information systems and computer

support. This IT support requirements has expanded with the advent of e-

commerce; utilizing B2B (Business to Business) and P2P (Partner to Partner) e-

commerce. There has been an increasing tendency to set up consortia that rep-

resent several players in a given field collaborating with one another to form a

large logistics consortium in order to form one organization to compete with

larger competitors and/or extend beyond their region of operation. This paper

deals with the management of collaborative workflow changes in such consor-

tia and the adaptation of these changes to existing workflow systems. We also

discuss issues of adaptation to new systems, workflow mining techniques for

adaptation and a proposed prototype to capture the meta-data to implement the

workflow system.

1 Introduction

This paper aims to develop the modeling approach for dynamic business processes for large

logistic consortia, where business strategies can be captured vigorously while simultaneously

allowing workflow changes to be handled. Often we see that the business processes are com-

posed of several parts, a structured operational part and an unstructured operational part, or

they could be composed of semi-structured parts with some given and some unknown details.

Unpredictable situations may occur as a result of changes by management. The inability to

deal with various changes greatly limits the applicability of workflow systems in real industrial

and commercial operations. This situation raises problems in workflow design and workflow

systems development. We propose workflow modeling and workflow mining techniques for

the adaptation of workflows to the existing workflow system

2 Collaborative workflow model for logistics consortium

The advent of the web to bind organizations together, for carrying out sales over great dis-

tances and at any time has created new modes for marketing and enabled partnerships, previ-

ously inconceivable within a wide array of businesses, as well as other human activities.

A consequence of this connectivity and information richness is that one is faced with an in-

creasingly dynamic business environment and marketplace.

Warehouse logistics are activities that require strong information systems and computer

support. This IT support has expanded with the advent of e-commerce. However, with the

advent of B2B (Business to Business) and P2P (Partner to Partner) e-commerce, there has been

Pudhota L., Chang E., Davis J. and Venable J. (2004).

Collaborative Workﬂow Management for Logistics Consortium.

In Proceedings of the 1st International Workshop on Computer Supported Activity Coordination, pages 246-252

DOI: 10.5220/0002675802460252

 SciTePress

an increasing tendency to set up consortia that represent several players in a given field. Such

consortia consist of companies or organizations in a given field that get together and produce a

single site or what appears to be single site in order to increase traffic through the site com-

pared to other competitor’s sites and/or extend beyond their region of operation.

2.1 General Organizational Workflow Models

In general, we found that most workflow modeling and workflow designs are only concen-

trated on the operational aspects of the organization [Siebert, R, Han Y et. al, Marshak, R.T].

Although operational aspects of workflow designs are crucial to the organization, we note that

they are passive in changes and are not dynamic. They only change when there is a manage-

ment decision. Collaborative workflow is a new type of workflow that has to be integrated into

the existing operational workflow.

2.2 Organizational Workflow Models

As per the paper by Marshak, R.T.: “Falling in Love with Distinctions”, in the “New Tools

for New Times: The Workflow Paradigm”, Future Strategies Inc., 1994 [Marshak, R.T] states,

in any organization, a workflow can be disseminated into three categories. These three types

of workflow are supported by three categories of workflow systems.

1) Ad-hoc workflows involve human coordination. For example, office processes such as

product documentation or sales proposals.

2) Administrative workflows involve repetitiveness. For example, routing an expense report

or travel request through an authorization process.

3) Production workflows (automated tasks being performed repeatedly) production work-

flow encompasses an information process often involving interaction with one or more dis-

tributed/ heterogeneous/autonomous information system.

According to Dr Hala Skaf-molli [http://www.loria.fr], workflow can be categorized into

four groups:

1) Collaborative workflow

2) Production workflow

3) Administrative workflow

4) Ad-hoc workflow

2.3 MAO Model [Managerial, Administrative and Operational]

In our study, we have found that there are many workflow components within an organization

and these workflow components interact with each other to achieve the organizational goals

and objectives. Therefore, we disseminate the organization’s workflow into the following

levels:

a. Operational workflow relates to the core business operations. It is usually measured by its

performance and by the volume of its output. The operational workflow is the main source

of value generation for the organization.

b. Administrative control workflows are involved in making decisions and the prioritizing

and scheduling of tasks. The administrative task workflow is measured by its efficiency.

c. Managerial workflow carries out business decisions, which in turn control entire business

administration and is measured by the financial and final results.

Here, we see that customers, suppliers and partners are external entities to the organization’s

administrative component. Alternatively, they can be partly out of the actual organization in

the form of branches. Similarly, operational components can be outside the organization in the

form of a partner possibly in an overseas location.

247

2.4 Collaborative MAO Model

The advent of the Internet has provided mechanisms to allow organizations to bind together for

carrying out sales over great distances at any time. It has created new modes for operation for

service and marketing and enabled partnerships, previously inconceivable, within a wide array

of businesses as well as other human activities. A consequence of this connectivity and infor-

mation richness is that one is faced with an increasingly dynamic business environment and

workflow. Several factors characterize this collaboration [Chang, E. et. al 2000], namely:

a. A strong information infrastructure that extends beyond the original closed walls of the

individual enterprise.

b. High connectivity and electronic handling of information of all sorts including data and

documents.

c. An increasingly collaborative approach between what were more traditional, individual

enterprises.

d. Utilization of new forms of electronic interaction, provision of services and utilization of

services.

e. Ability to self-organize and reconfigure the business of the organization; perhaps even the

organization as a whole.

f. Use of multiple channels for sales and marketing.

A key factor in the success of such collaboration is the creation of the underpinning infor-

mation infrastructure to carry out the required services and development to enable and support

the creation and the strengthening of small-medium enterprises (SMEs) to achieve some of the

characteristics of collaboration. We have found that our MAO model (refer to figure 1) is

suitable for a collaborative environment.

Fig 1. Collaborative Organizational Workflow

3 Issues in Adoption of New Systems

An organizational workflow is a composition of a number of workflow components. Some

workflow components are active when changes take place and some are stationary.

So far, little has been addressed in literature in the classification of workflow components,

their inter-relationships or formal definition of their attributes and processes. There has been

some existing workflow modeling techniques used in literature to help model the workflow and

dynamic aspect of the workflow, such as Petri Nets [Aalst, W.M.P et al,1999, 1996, Jensen, K

Managerial

Administrative

Administrative Administrative

Op Op Op Op Op

Customers

Suppliers

Partners

Organisation

Managerial

Administrative

Customers

Suppliers

Partners

Organisation

Managerial

Administrative

Administrative Administrat ive

Op Op Op Op Op

Customers

Suppliers

Partners

Organisation

248

1992], Event-driven Models, State Event and Action Rules [Nutt G 1996, Aalst, W.M.P van

der; Jablonski, S 2000, Joeris, G 2000, Ortner, W., Starty C, 1999], UML Activity Diagrams,

Sequence Diagrams and Extended Activity Diagrams [Gautama, E Chang, E. 2001, Stary, C

1996, Gantama, E.,et al, 2003]. We found that these techniques allow for modeling of existing

processes. However:

a. They do not model the inter-relationship between the workflow components. Some aspect

of concurrency and asynchronous execution of the different workflow process (i.e. multi-

threading);

b. Their current use does not deal with the dynamic aspect of workflow models or provide a

clear indication where the flexibility is allowed, when changes occur or how the organiza-

tion can adapt the changes at Just-In-Time; also

c. These modeling techniques only model one aspect of the organizational workflow and

sometimes have too low a level of representation and they are basically not applicable at

the conceptual level of development of complex organizational workflow systems [Bos-

sidy, L. & Charan, R. 2003, Mohan C 1999, Chang, E. et. al 2003].The necessity is for

modeling the workflow at a higher level of granularity that involves many sub-workflow

components and workflow processes, their interfaces, interaction and relationships. Man-

agement is often not in a position to make realistic assessment of whether their organiza-

tion can execute the decisions taken by their collaborative consortiums and if these deci-

sions can be implemented in the workflows at the production level.

In this paper, we propose using workflow mining to support the transfer of this unstructured

information and also to determine if these decisions are valid enough for workflow implemen-

tation, thus helping management to get feedback on their decisions in the better running of the

organization.

4 Workflow mining technique for Adaptation

Fig 2. Workflow Mining Strategy for Logistics Consortium

The decisions taken by the consortium board, which is represented as unstructured data is

passed on to the collaborative consortium management. This data which is unstructured is

converted into structured data through a process of unstructured data management (UDM), a

249

process of mining, organizing and analysis to extract actionable information. Here, UDM is

used as a text mining tool to convert the unstructured information into structured data by con-

sidering precision, thoroughness and relevance of the unstructured information provided

(please refer to figure 2).

Creating workflow processes is complicated and time consuming as per the requirements of

management. To support the continuous design process, we propose using a workflow log

which contains information about the workflow processes, information collected at runtime and

can be used in the diagnosis and redesign phases to derive a model explaining the events re-

corded in workflow mining. Internal factors are passed on for further consideration in the

decision making processes by the management; otherwise it is a cyclic process in creating a

new workflow model from existing workflow.

5 Prototype Implementation

We propose using a meta-data based application generator for dynamic workflow system man-

agement. This meta-data application generator [John Davis et al 2004] contains meta-data

definer, run-time updater, application meta-data and a run-time processor. Meta-Data Definer

is made up of a meta-data design editor which provides efficient methods for defining the

meta-data, Third party design import wizards to import third party design toolsets and convert

them into corresponding meta-data syntax and a data source reverse engineering wizards to

reverse engineer existing data base schemas and convert this schema to corresponding data

dictionary meta-data to accelerate the meta-data system design process.

Run-time updater updates live program and data dictionary meta-data for existing applica-

tions. Application meta-data consists of platform independent application meta data. Run-time

processor consists of platform dependent visual component mappings, event processing map-

ping and DBMS mapping. Here, mapping of meta-data takes place for various structures,

components and data schemas and in the final stage, we propose using some level of platform

dependent drivers to generate a workflow.

A workflow model represents a group of workflow components. It explicitly captures dif-

ferent characteristics of the target area. Frequently, the choice of characteristics to retain is

dictated by relevance [OMG 1999]. We propose the steps of modeling the workflow which

can result in effectively capturing the changes

Fig 3: Conceptual View of Meta-data Driven Architecture for Collaborative Management

250

Requirements from customers are changing and functionalities have to be added to the

software already created, which may require new technologies. Therefore, to enable the inte-

gration of the old system with the new system and any other system created in the future, meta-

data driven architecture has to be used.

Meta-data driven architecture includes the iterations of deploying, combining and managing

data as well as applications. Meta-data driven architecture is made up of a platform-

independent UML model and one or more potentially platform-specific models. In addition,

the platform independent model contains information about how the UML base model is im-

plemented on the middleware platform.

Meta-data driven architecture utilize Model Driven Architecture (MDA) concepts [OMG

1999]. Currently there are at least 40 tools that incorporate at least one major aspect of MDA:

UML based modeling transformation between the applications over all design models and the

models that are specific to the underlying computing architecture (.NET, EJB and so on); and

the generation of code in a specific language. Williams estimates that MDA generates about

50% to 60% of the application code but it also depends on applications, using this approach we

hope to generate a new workflow system.

6 Conclusion

In this paper, we have discussed the issues faced in collaborative organizations, the develop-

ment of workflow systems based on decisions taken at managerial levels and the adaptation of

these issues in new workflow systems. We have proposed using workflow mining technique

for adaptation to existing workflow and introduced the concept of workflow mining and work-

flow log, where we need to work on the issues and their further role towards adaptation into the

workflow system.

References

Siebert, R 1996..: Adaptive workflow for the German Public Administration. International Conference on

Practical Aspects of Knowledge Management,

Aalst, W.M.P. van der; Basten T.; Verbeek, H.M.W.; Verkoulen, P.A.C.; Voorhoevoe, M March 1999..:

Adaptive Workflow On the Interplay Between Flexibility and Support. Proceedings of the first Interna-

tional Conference on Enterprise Information Systems, Vol 2, pages 353-360, Setubal, Portugal,

Han Y.; She th, A; Bussler,C. , 1998: A Taxanomy of Adaptive Workflow Management, Proceedings of the

CSCW-98 Workshop Towards Adaptive Workflow Systems, Seattle, USA, November 14.

Nutt G 1996..: The evolutions Toward Flexible Workflow Systems, Distributed Systems Engineering, Vol.

3, No. 4, pp. 276-294, December

Aalst, W.M.P van der; Jablonski, S 2000.: Dealing with workflow change: Identification of issues and

solutions, Special

Issue of International Journal of Computer Systems Science and Engineering, September.

Joeris, G 2000..: Modeling of Flexible Workflows and Their Decentralized Enactment in flow.net, Special

Issue of International Journal of Computer Systems Science And Engineering,

Ortner, W., Starty C, 1999.: Virtualization Of Organisations: Consequences for Workflow Modeling,

Proceedings of the 32

Hawaii International Conference on System Sciences.

Stary, C 1996.: Integrating Workflow Representation into User Interface Design Representation, Software –

Concepts and Tools, Springer,.

Object Management Group. OMG UML 1.3, June 1999. OMG doc ad/99-06-08.

Mohan C 1999.:Tutorial: State of the Art in Workflow Management System Research and Products”, ACM

SIGMOD International Conference on Management of Data, Montreal, Canada, June

Marshak, R.T 1994.: “Falling in Love with Distinctions”, In “New Tools for New Times: The Workflow

Paradigm”, Future Strategies Inc.,

251

Aalst, W.M.P. van der 1997: “Verification of Workflow Nets”, In P. Azema and G. Balbo, editors, Applica-

tion and Theory of Petri Nets, Vol. 1248 of Lecture Notes in Computer Science, pp. 407-426, Springer-

Verlag, Berlin,.

Aalst, W.M.P. van der 1998: “The Application of Petri Nets to Workflow Management”, The Journal of

Circuits, Systems and Computers, 8(1):21-66,.

Jensen, K., 1992 “Coloured Petri Nets: Basic Concepts, Analysis Methods and Practical Use. Volume 1:

Basic Concepts”, EATCS Monographs on Theoretical Computer Science, Springer-Verlag, New York,

Gautama, E Chang, E. 2001 Isorc 2001 IEEE conference on Distributed Extended Activity diagram for

workflow modeling. Germany July.

http://www.loria.fr/~skaf/cours/workflow/workflow1/sld001.htm

Gantama, E., Chang, E., Jayaratna, N. & Pudhota, L., 2003 “ A methodology for flexible work flow

modeling “ Int. Journal on Computer System, Science and Engineering, Vol 18 No3,UK.

Bossidy, L. & Charan, R. 2003, Execution, New York

Chang, E. et. al 2003. web services for collaborative logistics IEEE conference on industrial

informatics Benff, Canada Sept.

John Davis, Andrew Tierney, E Chang 2004 “Meta-data frame work for enterprise information

systems specification” ICEIS Portugal

252