Knowledge-based flexible workflow to support decision follow-ups

The importance of improving the quality of decisions in organizations has been a subject for several years in computer, management and decision sciences. Some results were reached, but still a lot of questions remain open. From the perspective of computer science, systems like DSS, GDSS and ODSS are examples of those results. This research proposes another approach, which is based on the concept of a post-decision phase and the technological support that could bring contributions for the decisionmaking process. Through this technological support we aim that part of the process that takes place during a post-decision implementation can contribute to improve the quality of decisions made within an organization. Some innovative aspects include the timing where it is supposed to be used, i.e., after a decision is made and also, it considers news perspectives as a combination of a flexible workflow system integrated to a knowledge repository. Introduction Presently, organizational scenarios demand a lot of interaction among team members to solve problems and daily processes. Independent of the hierarchical level, employees need to exchange information, to make decisions, to obtain information from different resources, to develop projects, to sale products and services, counting gradually more on group collaboration. Nevertheless, the existing tools proposed up to this moment solve only part of this complex scenario. Examples of these tools are e-mail, videoconferencing, instant messengers and even text editors. But we can argue, why these groupware tools are not sufficient? One reason is the lack of integration among them. There are many tools available, but one does not exchange information with the other. This work has to be done by the user, most of the time. Another point to be considered in organizational scenarios is the different demand that each group of user has related to technological supports. The great majority of successful tools available in the market aim to solve problems for the employees of low hierarchical levels within organizations. There is a lack of tools to support the upper management demands. Some successful tools were already proposed and are under an evolution process. Examples of these tools are Data Warehousing Systems, Datamart and Executive Information Systems. But, a common characteristic of these tools is the operational data dependency, which has a great value, but it is far to be enough. There is still a big gap to be filled considering other types of data, which also have value for the organization, as decision related systems, knowledge management tools and organizational learning support tools. In spite of the big gap to solve organizational problems related to upper management users, this research, does not aim to be the solution for every problem. The focus of this proposal is on the process that takes place after a decision is made, referred here as “post-decision” phase. The approach is to use a technological solution to support the documentation of a decision made, a formalization of the plans related to its implementation, a systematic control of the consequent tasks, the people involved, and documents generated, so that a knowledge repository is built helping the organization to improve new decisions, identifying indicators for best-practices, as well as enabling an organization learning process. The next section presents the problem considered and the hypotheses for this research. Then, comes the proposed solution, with its respective information about the target audience. Further comes a section describing some related research and a list of commercial tools to provide a comparative view. The last two sections present, respectively, the challenges involved and the validation techniques considered. In the end of the paper the next steps of this research are presented. The Problem Problem’s Context One problem that has been mentioned for a long time in several distinct researches is the scenario of decision-making process. How shall a decisionmaking be more effective, involving teams, complex problems and its consequent implementations? Problems related to capture, measure and improve the quality of decisions have been cited for several decades. Since the 80’s authors suggest this complexity. For example, DeSanctis in 1987 makes a reference to previous researches that say, “our society is experiencing the emergence of a post-industrial environment characterized by greater knowledge, complexity, and turbulence” (cited by Huber 1984b) and “one effect of this trend is that decision-related meetings are becoming more frequent and more important. At the same time, the decisions confronting groups are becoming more complex and must be more quickly, and with greater participation than in the past” (cited by Huber 1986), see (DeSanctis, 1987). And still, this scenario has not changed expressively during these 18 years, in spite of the several initiatives proposed during the past years. Some examples can be found in (DeSanctis,1987), (Keen, 1987), (Sridhar, 1998), (Power, 1998) and (Power, 1999). Another problem, which is also an open problem, is how to provide a follow-up of a decision-made implementation. The gap between the time a decision is made and its corresponding post-decision implementations may, in fact, turn several decisions inconsequent, due to the lack of appropriate support to the implementation follow-up. Post-decision activities have not had enough research within the decision-making cycle. Perhaps they have been considered trivial or not meaningful in the past. However, without an appropriate follow-up, important decisions made in the previous phase may get lost or be implemented wrongly. Many researchers are interested in what happen during the process of decision making, but what happen after the final decision and its consequent steps? How are they related? How they can contribute to improve organizational processes or to help in new decisions? My proposal is to capture and organize the information and knowledge, involved during the post-decision implementation, and make them available, so that new decisions can avoid to committing past mistakes and best practices can be discovered. Problem Description and Related Hypotheses The problem considered is: how one can improve the quality of decisions in an organization? Several researches were and still are done in the direction of capturing the rationale generated during meetings and interactions in a decisionmaking process. This research considers another perspective, which is focused on the following steps after a decision is made. We think that the information and knowledge generated during a decision implementation, if captured, contextualized and put on disposal, can also contribute to improve the quality of decisions and to show directions for best practices. This proposal is based on a set of technologies with seem to be adequate to contribute to the solution of this problem. Table 1 shows a schematic relation between three organizational scenarios based on citations of the related literature (see Ramesh, 2001; Stewart, 1997; Markulla, 1999; and O’Leary, 1998) and three hypotheses considered in the scope of this research. Organizational Scenario Hypothesis Many mistakes are frequent in organizations and some of them are related to the lack of information and knowledge about previous experiences; Learning with past mistakes and successful stories can improve organizational practices, organizational learning and to avoid recurrent mistakes; Other mistakes are frequent because of the lack of planning and control of decision implementations; Formalizing a decision implementation, using a workflow management system, allows planning, tracking and getting warning signals about problems, providing control over a decision implementation; Knowledge and information are generated during a decision implementation; The use of technology in organizing, storing and making available information and knowledge generated during a decision implementation can contribute to improve the quality of future decisions; Table 1: Organizational scenarios and related hypotheses The Proposed Solution The explanation about the solution will follow ann organizational cycle that takes place after a decision is made and its relation with a knowledge repository (figure 1). Step1 is related to the timing a decision is made. After that, the creation of an expected plan should be done, which is represented by step 2. Then, on step 3, an execution plan can be implemented and finally another phase should take place to evaluate the whole process, represented by step 4. Observe that these steps are not necessarily following a sequential line. For e.g. during the execution of a plan, maybe it is necessary to go steps backward and make a new decision, or during the execution of a plan it can be necessary to go backwards and make a new plan, or split the original plan into two different plans. Still considering figure 1, the rectangle that represents the knowledge repository has a different orientation, because it is a feature that can take place at any time within the cycle. For each different step the use of the knowledge repository can be served as input (for e.g. the description of a decision made on step1) or output (for e.g. recovery of data related to similar decisions in the past helping in creating a plan to a new decision).