DECISION FOLLOW-UP SUPPORT MECHANISM BASED ON

ASYNCHRONOUS COMMUNICATION

Carla Valle, Wolfgang Prinz

Fraunhofer Institut - FIT, Schloss Birlinghoven, Sankt Augustin, 53754 - Germany

eywords: Decision Support Systems, Asynchronous Communication

Abstract: Decision management and decision support systems are themes under investigation for several decades, and

both research areas provided contributions for the quality of decision making processes. However, little

work has been done in the aerea of decision follow-up, especially regarding decisions made during

meetings. In this paper we analyse the concepts related to this problem and we propose a solution based on

mechanisms supported by computer to assist the formalization of meeting outcomes, and to provide decision

follow-up.

1 INTRODUCTION

One important area vastly investigated in the

literature is the area of Decision Support Systems

(DSS). The main goal of these researches is to

improve the decision making process through the

use of this technology.

One of its sub-areas is related to the follow-up

process of decisions made. Once a decision is made,

there is a lack of information about how its

implementation take place, who is involved and

making use of which resources and what are the

current problems. As a consequence to this fact,

decisions are wrongly implemented or not

implemented at all. Often, decisions that are

implemented without the necessary follow-up may

generate outcomes, different from those planned at

the time of the decision. Besides that, cultural

barriers and the lack of appropriate tools induce just

informal links. As a result, important decisions are

not properly or timely implemented (Borges, 2002).

The mechanisms used nowadays to provide

decision follow-up are more based on user efforts

than on system efforts, like for example, informal

conversations, free emails, project management

tools, workflow systems and simple to-do list tools.

However, a lot of data is constantly created by users,

either through documents, electronic

communication, phone calls, reports, and so on,

which could be used to automatically provide some

levels of follow-up to decision makers and to

decision implementers.

Several initiatives were and still are done in the

direction of capturing the rationale generated during

meetings and interactions in a decision-making

process (Keen,1987), (Kraemer,1988),

(Kleidorfer,1993), (Santhanam,2000). This research

considers another perspective, which is focused on

the steps that take place after decisions are made

during meetings. Our idea is based on the

formalization of meeting outcomes with decisions

made explicit and a set of automatic mechanisms to

analyse, categorize and provide awareness

information based on email content analysis related

to the decisions made.

With these ideas we aim at improving the

communication between decision makers and

meeting participants with those people who really

implement the decisions. We also consider that these

types of automatic support can improve the track of

decisions using the data users already generate

"naturally", in order to improve decision quality, but

also to make users aware of problems and decision’s

implementation.

This paper is organized in the following way:

section 2 shows some related work, section 3 shows

the problems handled in this research, section 4

presents the concepts used in this research, section 5

presents a set of mechanisms identified to solve

problems presented in section 3 and, finally, in

section 6 we raise some conclusions up to this

moment.

423

Valle C. and Prinz W. (2004).

DECISION FOLLOW-UP SUPPORT MECHANISM BASED ON ASYNCHRONOUS COMMUNICATION.

In Proceedings of the Sixth International Conference on Enterprise Information Systems, pages 423-429

DOI: 10.5220/0002654504230429

 SciTePress

2 RELATED WORK

In (Wiberg, 2001) the author presents a mobile

physical/virtual meeting support system intended to

support knowledge management (KM) in mobile

CSCW, named RoamWare. One of the objectives of

the project was to support the transformation from

socialization to externalization so that informal

meetings could be made explicit and used in

knowledge processes. The Roamware system is

designed to support people who are moving around

and participate in different mobile meetings. The

system is intended to be used in the background

while walking around, collecting data about other

co-located persons who also have the system

running. The idea is that the collected information

will be used in later setting when the users are

geographically dispersed from each other.

In (Borges, 2002) the authors discuss the use of a

post-meeting support workflow-like solution where

working plans can be described and enacted. The

proposed solution is a system combining a process

design tool with a workflow enactment tool.

However, given the ad-hoc nature of the processes

described in the paper, the authors suggest that a

commercial Workflow Management System

(WfMS) alone would not be enough to support the

requirements of the scenario of post-meeting phases.

Even with the adoption of a WfMS for process

enactment, additional monitoring tools are proposed.

In (Costa, 2000) the authors propose a solution

for supporting meeting report processes, based on

the notion of genre and systems of genres. The idea

of this system is to support and help meeting

participants in the process of creating and

dissemination meeting reports. The system is

composed by two tools, an analysis tool, which

helps users identify meeting genres and a reporting

tool that guides users in the production of meeting

minutes.

All these approaches are valid and have

considered different perspectives of the complex

problem of providing decision follow-up related to

meetings. In the following sections we provide more

details regarding our solution.

3 DECISION FOLLOW-UP AND

MEETINGS

Our proposal looks at decisions made only in

meetings, since meetings are one of the most

common practices where decisions are made

(Simon,1966), (Stefik,1987), (Nunamaker,1997).

Our decision follow-up mechanism is going to be

related to post-meeting activities and our starting

point will be the meeting minutes with decisions

made explicit.

Decision meetings are not isolated events. They

are part of a continuous cycle of premeeting,

meeting and post-meeting activities (Bostrom,1993).

The meeting itself is the most visible part of this

cycle, but the other components are always present.

Making premeeting and post-meeting activities

explicit may be the first step to enhance the whole

cycle and thus, to obtain better decisions as a final

result. All the three phases can be considered equally

important, since they deal with different aspects of a

decision. Nevertheless, few tools have been

proposed to support premeetings and post-meeting

phases. In (Borges,1999) the authors detail the

premeeting phase and its relation to meetings. While

in (Hayne,1999) and (Nunamaker,1997) the authors

discuss the meeting particularities.

The post-meeting phase is when the

implementation of the decisions is executed. This

stage contains activities to be carried out by people

not necessarily present in the meetings. Its activities

include dissemination, monitoring implementation

of the decisions and clarification of ambiguous

decision details. We believe this phase involves

knowledge that can be useful in future instances of

the cycle. Experiences of the implementation phase

can provide indicators for new meetings, and stories

about what was implemented with or without

success (Valle,2002). And last, but not least, new

decisions may appear during this phase. These

decisions may or may not be related to other

decisions made, but they should be observed and

checked as those related to meetings.

Decision follow-up can be made through several

management tools (e.g. project management tool,

workflow systems, intranets, document management

system), depending on cultural, organizational and

financial reasons one or a combination of them can

be used. But there is a common denominator for

almost any modern organization, which is one of our

beliefs to provide decision follow-up: the electronic

communication. Communication is one of the richest

components of the whole decision making process.

It is through communication channels that people

exchange knowledge and opinions, raise problems

and solve them. Our proposal is based on the idea

that formalizing decisions in an electronic format

should be the first step to provide decision follow-

up; and the second step should be the analysis of

related electronic conversation.

Communication can be realized in different ways

(synchronously, asynchronously) using different

channels (text, voice and images) and supported by

various tools (telephone, fax, email, chat, discussion

forums, videoconference, voice conference, etc).

ICEIS 2004 - ARTIFICIAL INTELLIGENCE AND DECISION SUPPORT SYSTEMS

424

In decision follow-up, communication seems to

be one of the most important aspects to be observed.

Actually, formal or informal communication are

used to transmit information an knowledge, to make

new decisions, to externalize problems and

successful stories, and so on. Since we aim at

discussing how computers can support decision

follow-ups, especially based on the information

available in communication processes, our analysis

will be restricted to the possibilities supported by

computers.

4 DECISION FOLLOW-UP BASED

ON ASYNCHRONOUS

COMMUNICATION

Many users rely on emails to get up to date

information about projects, tasks, responsibilities

and track of on going processes. Email has great

advantages over other tools. It is simple, not costly

(except the storage costs, which are becoming very

high), it is already accepted as a communication tool

in most organizations and it supports asynchronous

communication. On the other hand, e-mails are not

easily structured. Some available tools provide basic

mechanism to organize them in folders, sub-groups,

by sender/receiver, incoming and outcoming

messages, but the context of the messages and

mostly their contents are not analysed by these

mechanisms.

In the case of decision follow-up, emails can be

very useful. Many project leaders, mainly of

distributed projects, rely on email to catch up with

information. Despite the information being there, its

content analysis has to be made by users. If a user

really desires to structure a conversation, not only

based on the "subject" field of the email messages,

the user would have to read each message again and

create "artificial" structures to store them in a

meaningful way.

This proposal aims at building a combination of

mechanisms to provide decision follow-up looking

at asynchronous communication based on e-mails.

One reason for that choice is based on the fact that e-

mails are one of the most successful electronic

communication tools, with a stable utilization across

organizations (Levitt,2003). In (Wired,2003), the

study done shows that for power users, typically

better educated and higher earners (managers), it is

necessary to spend about 2 hours or more daily on e-

mail, often beyond four, not only because of their

direct participation in the communication process,

but also as a person who receives copies of others'

emails to be aware about on-going projects they are

involved in.

5 SYSTEM DESCRIPTION

The mechanisms proposed aims at providing four

main groups of functionality: a formal

documentation of decisions made, through the

outcomes of meetings; the content analysis of

messages related to decisions made; the

categorization of messages related to decisions

made; and awareness (notification and categorized

messages) about the follow-up process of decisions

made (Figure 1).

Formal Documentation of Meetings: We are

proposing for the starting point of decision follow-

up, a mechanism to formalize the decisions or to

make them explicit. We propose the use of a special

type of email, based on a template, where a meeting

participant writes the minutes in a semi-formatted

Meeting

Minutes

Mail Server

Content Analysis

Mechanims

Categorization

Mechanism

Awareness

Mechanism

Decision Follow

Mechanism

Emails

Figure 1: Components of the decision follow-up

mechanism

DECISION FOLLOW-UP SUPPORT MECHANISM BASED ON ASYNCHRONOUS COMMUNICATION

425

way, pointing out decisions made which are going to

be used to trigger the decision follow-up process.

The filling out action of this form should be done,

ideally, during a meeting execution. The fields

proposed in this email template are supposed to

represent different categories of information. Firstly,

general information should be provided, followed by

the objectives of the meeting and the name of the

attendees, or participants.

During the meeting activity, the noter should list

each topic discussed, followed by any sub-topic,

plus the details about the "to-do-list", and who is

going to take the responsibility, when the actions

should be done. And if applied, the noter could

classify the type of topic/sub-topic and define if it

should be followed-up or not. The template should

contain 1 or N topics with 1 or N sub-topics. The

number of them will be determined during the

meeting execution. For each topic or sub-topic we

propose the use of a "type" field, which will have as

initial values: decision, idea/suggestion, simple

presentation of a topic, simple comment,

explanation, question, answer, not applicable; and

other (with blank field).

Considering a participant of the meeting will

execute the activity of taking notes and filling out

the email template, the decisions would somehow

become explicit, either in the explicit classification

of topics and sub-topics, or in the "to-do-lists" or a

combination of them.

At the end of the annotation process, a first parse

could be made to create a summary of what was

considered decision, what was not and ask for user

confirmation. For this first parse tool, we consider to

use the explicit fields values content analysis.

We would like to emphasize that in a real

situation this effort of classifying meeting topics as

decisions to be followed-up, may be more adequate

to decisions regarding innovative ideas, projects, or

decisions that demand high level interaction among

different people, to decisions that offer high critical

degree of implementation, or to decisions considered

risky.

We are also considering that at least one

responsible for each decision made must be

assigned. This is the first user to be informed about a

decision follow-up.

After the minutes are ready, the message should

be sent via email to participants and to all people

who should be informed about decisions, actions and

needs.

Tracking of Messages: After the first step is done,

with the triggering minutes email containing

decisions made, users will start to send messages to

each other related to those decisions. At this moment

we need mechanisms to check the existence of new

messages and mechanisms to categorize them,

according to their nature (E.g. new decisions,

problems related to decisions, new meetings, etc).

For this phase we plan to have agents working at

the mail server side, analysing the entrance of new

messages and their contents. It is predictable to be

necessary to develop the following agents:

- A content analysis agent: this agent will be

responsible to analyse each new message and

separate them as a new input to the decision

follow-up mechanism (E.g. message containing

meeting minutes), or follow-up messages (E.g.

messages related to meeting minutes or to the

decisions made contained in any meeting

minute), and finally, not related messages (E.g.

messages out of context, or private messages).

- A categorization agent: after the first agent has

done its work of filtering the different types of

messages, a second analysis has to be executed

to categorize the messages by another agent.

This agent is supposed to create virtual links

between messages, so that decision follow-up

categorizations can be generated. Every

message has to pass through this analysis and

the agent will try to relate the message to one or

more decisions made.

At this phase of the research, it is still not clear if

it is going to be possible to allow free-mail text or if

the messages in the context of decisions should be

semi-structured to make the analysis made by the

agents feasible and meaningful. Ideally, we would

prefer to allow free-text.

Presently, there are several tools available to

help us analyse the content and the categorization of

messages (Xelda,2003), (Levitt,2003) and (Protan,

2003) are some examples. We plan to analyse a few

of them and choose one for using in this research

with the required adaptation implemented to support

our objectives.

Awareness Mechanisms: as the final step of our

proposal, we want to develop awareness

mechanisms to provide the follow-up of decisions to

those who are involved or interested in obtaining

information about decisions made.

A decision can affect many people besides those

directly involved in its conception and

implementation. It may be of peripheral interest, for

example, to high-level management. On the

contrary, a decision will be of direct interest to

people affected by it (Borges,2002). One way of

providing the right information to the right person at

the right time is through the use of awareness

mechanism.

ICEIS 2004 - ARTIFICIAL INTELLIGENCE AND DECISION SUPPORT SYSTEMS

426

Considering that not every detail about a

decision implementation is relevant to all

participants or related people, the automatic

mechanism should somehow filter the information to

provide the right awareness. For the problem dealt in

this research, we plan to develop the following

awareness mechanisms:

- Awareness about new decisions: every time a

new decision is made, people related to it

should be notified. Based on our proposal on

formalizing meeting minutes, the message

where the meeting minutes are described will

solve this first awareness need.

- Awareness about a decision follow-up: every

time a message is sent to the system, after being

analyzed, the system should provide awareness

about the existence of a "follow-up" message

related to a decision, if that is the case. At this

moment, we think to use as content criteria,

keywords related to the decisions made. The

criteria used by the system to determine to

whom it should send the follow-up is defined at

the meeting minutes participants or involved

people, combined with user preferences. The

timing of receiving this awareness should also

be user configurable. Some users may want to

constantly receive information, while others

may prefer to see them in a daily or weekly

format, just to cite some possibilities. We plan

to develop different ways of visualizing the

follow-up, for example, in threaded messages or

in graphic representations, again depending on

users preferences.

- Specific awareness based on decision life-cycle:

we plan to provide slightly different types of

awareness according to the decision life-cycle:

new decisions, problem related to a decision,

decision being changed, decision being aborted,

decision implementation ended, etc.

Besides the pre-determined awareness

mechanism presented, the users should be able to

query the message base in order to get historical

information about decisions made and their

respective follow-up, at any time.

It is also important to provide users with the

possibility of aborting the decision follow-up. In this

case, the responsible for stopping the specific

service should inform related people, for example

with a message, the justification for the cancellation.

Or this "stop" mechanism can occur every time a

decision is finalized, but in any case we think about

using a human intervention for this procedure.

User interaction modes: we aim at having different

user interaction possibilities:

- Meeting interaction mode: this is the mode

where users will input data about decisions and

meeting existence.

- Email interaction mode: this is the mode where

users will send emails related to decisions. We

are implementing this mode as an extension of

an email client. The basic idea behind that is

everytime a user wants to send an email, he can

decide whether to send it with a low, medium or

high level of contextualization regarding a

decision being implemented. The user can

choose to simply send a free message, which

should not be parsed by the content analysis and

categorization mechanisms, or he can explicit

click on a check box, thus enabling the decision

follow-up mechanisms, and allowing his email

for being used (or not) as a follow-up to others.

- Follow-up interaction mode: this is the mode

where the user will receive follow-up

notification and will be able to have different

ways of visualizing it. Again, this is being

implemented as part of an email client, as an

extension. This mode includes also functionality

to enable users to configure their follow-up

profiles, i.e., a profile where the user is going to

define the rules he wants the system to follow to

provide him with follow-ups. This includes the

timing and format configuration the follow-up

should be provided.

5 CONCLUSION

At this moment, many issues considered in this

research are still open. We aim at solving them

along the research activities so that we can provide a

solution to check the improvements on the decision

follow-up area. We are already conscious,

beforehand, of some drawbacks predicted up to now.

We will consider them during the activities, but we

are not sure if they will be part of the research

solution.

The first impact predicted from this proposal is

the way people organize meeting minutes. Following

the idea of creating the minutes during the meetings,

make users use a form-based template, supported by

a computer or similar (e.g. handheld) to document

the outcomes and decisions. If it occurs of not, the

other predicted impact is related to the way users

DECISION FOLLOW-UP SUPPORT MECHANISM BASED ON ASYNCHRONOUS COMMUNICATION

427

will organize the notes or even, the way meetings

organization might be affected. We expect that with

the proposed formalization, meetings become more

documented as well as the decisions made. On the

other hand, they can also start suffering from being

too formal and organized.

Another predicted impact is the use of email in

this scenario. Somehow we are changing the natural

and comfortable way users use this tool. As a

consequence of this proposal, we will have a social

impact about the use of email tools. If we choose a

"democratic" approach to send, analyze and receive

notification messages, where every message is

public, maybe users will feel constrained and so will

be the information written in the messages. Users

may not feel comfortable to have messages being

parsed and used to provide follow-up to others. On

the other hand, this change is just a matter of getting

aware about a technology, since even today all

messages running in organizational environments

can be stored, parsed and categorized if this is the

company policy. We also count on the idea that

giving and receiving feedback can bring a socially

related motivating and awarding perspective.

Clearly, this proposal falls in the group of

applications or mechanisms that need to be

supported by high administration levels at

organization in order to make people understand

their usefullness and limitations.

Another question to be solved is the number of

people who should be able to receive decision

follow-up. Since we aim at improving the

communication between decision makers and

implementers. In this case, the number of people

should be as big as users want or need.

Related to this topic, the overload of messages,

decision follow-up or awareness has to be

considered as well. An upper level manager would

be, in this situation, a candidate to receiving follow-

up about everything in the organization or in the

group where s/he works at. But, actually, not all the

information, and not all decisions should be of

her/his interest.

Some positive results are also expected. We aim

at having at the end of this work improvement about

meeting' minutes communication, decision follow-

up and awareness about on-going problems.

Somehow we expect the proposed system should

give users a payoff justifying the load they might

have while interacting with it.

We believe that with the help provided by such

mechanisms, better decisions could be made and the

timing to react over problems could be reduced.

Looking at the related on going research, we see

two interesting approaches to be investigated. One is

the integration of this proposal to a DSS, in order to

combine the decision follow-up with the rationale

used to make each decision. Another approach can

be a research looking at the possibilities of re-use of

experiences gained through the decision

implementation process and how this decision

follow-up can be used as "lessons-learned".

This research is under development, and there

are currently two investigations taking place: one

related to the definition of the adequate form-based

minutes with decisions represented explicitly, and

the other the extension of the email client

capabilities to support the mechanisms proposed.

REFERENCES

Balasubramanian, P, et al., “Managing process knowledge

for decision support”, Decision Support System, 27

(1999), 145-162.

Basoglu, A. N., et al., “A Framework for a Web-based

Group Decision Support System”, Informing Science,

June 2001.

Borges, M., Pino, J.A., Fuller, D., Salgado, A.C., Key

issues in the design of an asynchronous system to

support meeting preparation, Decision Support

Systems, vol. 27 no. 3, 1999, p. 271-289.

Borges, M., Pino, J.A., Valle, C., "On the implementation

and follow-up of decisions" in Proceedings of DSIAge

2002, Cork, Ireland, pp. 366-376.

Bostrom, R.P., Successful application of communication

techniques to improve the systems development

process. Information and Management, 16, pp. 279-

295.

DeSanctis, G. & Gallupe, B., “A Foundation for Study of

Group Decision Support Systems”, Management

Science, 33 – 5 (1987), 589-609.

Dourish, P. , Bellotti, V., Awareness and coordination in

shared spaces. In: Proceedings of CSCW’92, ACM

Press, Toronto, Canada, November 1992, pp. 107-114

Gross, T. Towards Ubiquitous Awareness: the PRAVTA

Prototype. In Ninth Euromicros Workshop on Parallel

and Distributed Processing – PDP 2001 (Feb. 7-9,

Mantova, Italy). IEEE Computer Society Press, Los

Alamitos, CA, 2001. pp 139-146.

Gutwin, C, Greenberg, S., and Roseman, M. Workspace.

Short Paper: Supporting Awarenes of Others in

Groupware. In Conference Companion of the

Conference on Human Factors in Computing Systems

– CHI’ 96. Vancouver – Canada, ACM Press

Hayne, S. C. The Facilitators Perspective on Meetings

and Implications for Group Support Systems Design.

The Data Base for Advances in Information Systems.

Summer-Fall, 1999. Vol. 30. No. 3,4. pp. 72-91.

Intext - http://www.intext.de/eindex.html consulted in Feb

28th, 2003.

ICEIS 2004 - ARTIFICIAL INTELLIGENCE AND DECISION SUPPORT SYSTEMS

428

Keen, P.G.W., Decision Support Systems: The Next

Decade, Decision Support Systems, 3 (1987), pp. 253-

265.

Kleidorfer, P. et al., Decision sciences: an integrative

perspective, Cambridge University Press, UK, 1993.

Kraemer, K.L., King, J.L., „Computer-based systems for

Cooperative Work and Group Decision Making”,

ACM Computing Surveys, Vol 20., No. 2, June 1988,

pp. 115-146.

Levitt, M., Mahowald, R.P., "There should be More to

Collaboration than Email", an IDC White Paper

Sponsored by Groobe Networks. Consulted in Feb.

2003, http://www.groove.net/extras/beyondemail/idc/

Power, D.J., “Web-based Decision Support Systems”, Part

I and II. The online Executive Journal for Data-

Intensive Decision Support, August 18 and 25, 1998.

Vol 2. Nos. 33 and 34.

Nunamaker, J.F., and Briggs, R.O. Lessons From a Dozen

Years of Group Support Systems Research: A

Discussion of Lab and Field Findings. Journal of

Management Information Systems, 13, 3, 1997, pp.

163-205.

Protan, http://www.upso.ucl.ac.be/protan/protanae.html

consulted in Feb 23rd, 2003.

Santhanam, R., Guimaraes, T., George, J.F. An empirical

investigation of ODSS impact on individuals and

organizations. Decision Support Systems, 30 (2000),

pp. 51-72.

Simon, H.A. The New Science of Management Decision,

New York: Harper & Row, 1966.

Simone, C. and Divitini, M., Integrating Context to

Support Coordination: The CHAOS Project. Computer

Supported Cooperative Work, 1999, vol 8, pp. 239-

283. Kluwer Academic Publishers.

Shim, J.P et al., Past, present and future of decision

support technology. Decision Support Systems, 33

(2002), pp 111-126.

Stefik, M. Et al. Beyond the chalkboard: computer support

for collaboration and problem solving in meetings.

Communications of the ACM. January 1987, Vol. 30,

Number 1. pp 32-47.

Valle, C., Prinz, W., Borges, M., Generation of Group

Storytelling in Post-decision implementation process.

In Proceedings of the 7th Intern. Conf. on CSCW in

Design. Sept. 25-27, 2002, Rio de Janeiro, Brazil /

Shen, Weiming [Hrsg.], pp. 361- 367.

Valle, C., Raybourn, E., Prinz, W. and Borges, M., Group

Storytelling to Support Tacit Knowledge

Externalization. In Proceedings of HCI International

2003, Crete, Greece.

June 22 – 27 (to be published)

Wired, News - Study Refutes Email Myth

http://www.wired.com/news/technology/0,1282,56781,00.

html Consulted in Feb 7th, 2003.

Xelda - http://www.mkms.xerox.com consulted in March

4th, 2003.

DECISION FOLLOW-UP SUPPORT MECHANISM BASED ON ASYNCHRONOUS COMMUNICATION

429