|
Article Excerpt
Abstract
An enterprise information portal (EIP) is viewed as a knowledge community. Activity theory provides a framework to study such a community: members of an EIP conduct specific tasks that are assigned through a division of labor. Each member of an enterprise information portal can undergo three distinct types of learning processes: learning-by-investment, learning-by-doing, and learning-from-others. Through these three types of learning processes, each member achieves specialized knowledge that is related to his or her own task. Cumulative knowledge resulting from the learning processes is considered in terms of two distinct attributes: depth and breadth of knowledge. This paper formulates a mathematical model and defines the goal of an EIP member as maximizing the net benefits of knowledge resulting from individual investment and effort. Numerical examples are provided to analyze patterns of optimal investment and effort plans as well as the resulting accumulated knowledge. The results provide useful managerial implications. In business conditions characterized by high interest rates or high internal rate of returns, it is preferable for members to delay spending their resources for learning. Intensive investment and efforts to obtain knowledge are preferable when the discount rate of costs is high, when knowledge is durable, when the value of knowledge is high, when the initial level of knowledge is high, when the productivity of the learning process is high, and when sufficient knowledge is transferred from other members. On the other hand, the size of the EIP has a positive or negative effect depending on the attribute of knowledge and the productivity of learning processes. Further properties of the optimal decisions and learning processes are analyzed and discussed.
Keywords: Knowledge management, enterprise information portals, learning, activity theory
**********
Introduction
There has been extensive research on knowledge management related to information technology (Alavi and Leidner 2001; Becerra-Fernandez and Sabherwal 2001; Earl 2001; Grover and Davenport 2001), but relatively little attention has been given to learning processes that form the critical part of knowledge acquisition in knowledge portals. This study focuses on knowledge acquisition processes and the individual user's investment strategy to obtain the optimal benefit from specialized knowledge in enterprise information portals (EIP). An EIP, in this study, is defined as a knowledge portal whose main function is to assist its members in obtaining specialized knowledge through various learning processes (Dias 2001; Mack et al. 2001). Further, an EIP can be considered to be a knowledge community (Bakos and Brynjolfsson 1997; Strader et al. 1998) that is composed of employees in a single company or in multiple companies that have relationships with each other. This paper utilizes an activity theory framework (Kim et al. 2002) to put forward the concept that members of the EIP conduct specific tasks that are assigned through a division of labor. The paper proposes a model wherein, through three types of learning processes (i.e., learning-by-investment, learning-by-doing, and learning-from-others), members obtain the optimal amount of specialized knowledge that is related to their own tasks. The three learning processes are modeled to have different impacts on depth and breadth of knowledge. Each member of an EIP decides optimal investments and efforts in each learning process, taking account of various conditions such as discount rate of cost, internal rate of return, and decay rate of knowledge. This study analyzes a mathematical model to explain the impact of each condition on individual decisions on the investment in three learning processes (Rao et al. 1995).
This research is expected to contribute to the literature in two ways. First, solutions of the mathematical model identify possible optimal individual decisions for obtaining the maximum benefit of knowledge under the given environmental conditions. The optimal solutions drawn from the analysis may explicitly provide individuals with proper guidance to knowledge investment. Second, the proposed model refers to an individual. However, the conclusions are equally applicable to a firm or a group as knowledge is a non-rival good in that there is no loss in sharing (Foray 2004). The findings of this study suggest appropriate guidelines for an EIP design policy to facilitate knowledge activities of members and achieve organizational effectiveness.
EIP as a Knowledge Portal and Learning Processes
Enterprise information portals are of multiple forms, ranging from Internet-based data management tools that bring visibility to previously dormant data so that their users can compare, analyze, and share enterprise information (Kim et al. 2002) to a knowledge portal, which enables its users to obtain specialized knowledge that is related to their specific tasks (Dias 2001; Mack et al. 2001). The EIP in this paper is assumed to be one that delivers information to users who constitute a knowledge community (Chan and Chung 2002; Goff 2001). Knowledge is a good that is often cumulative. Many types of knowledge such as databases, research tools, or generic knowledge or even physics are strongly cumulative, while others like songs and poems are non-cumulative.
Knowledge production has therefore the potential to create a combinatorial explosion.... This is a good which can be used infinitely to produce other knowledge which in turn is non rival and cumulative (Foray 2004).
Kim et al. (2002) explain knowledge management activities in the context of EIP from an activity theory perspective. In their model, each member of an EIP acquires or transfers knowledge through social interaction based on communities of practice. Each member of an EIP has specialized knowledge to accomplish tasks at hand in the organizational structure, resulting in division of labor (Brown and Duguid 1998). Each member of an EIP also exchanges knowledge among members. This study applies the same concept of the EIP to explain learning processes by which members achieve specialized knowledge associated with tasks.
Activity theory explains the transformation process of an object to outcomes in terms of three core components (actor, community, and object), three mediating components (tool, rule, and division of labor), and their relationships (Bellamy 1996; Hasan and Gould 2001; Kuutti 1996). Within the activity theory perspective, an individual member participating in the EIP conducts three activities for knowledge acquisition. First, as a member of the EIP, the member needs to keep making investments in knowledge to increase his or her level of knowledge. A certain amount of resources and time must be consumed to search for new skills and techniques. The outcomes of this learning process are specialized knowledge, which is associated with the particular task assigned by the EIP to the member, and the acquired knowledge, which increases efficiency in accomplishing tasks. An actor acquires necessary knowledge regarding his or her division of labor for performing the task or object. In this study, this process is referred to as learning-by-investment.
Second, the member should accumulate specialized knowledge by doing tasks in the EIP. Each member of the EIP has specific tasks within the EIP organization. While the member carries out assigned tasks, he or she should keep accumulating specialized knowledge for these tasks. Members learn from experience with working on their own tasks in the EIP, and they accumulate specialized knowledge by receiving feedback from their own experiences. It is also important for the member to prevent formerly accumulated knowledge from being lost. Even though the learning-by-doing process may be considered routine, all members of the EIP are still obligated to make efforts at accumulating and maintaining specialized knowledge that will allow them to work more efficiently than before. From the activity perspective, the transformation process of object (task) to outcome is the key component for the second process of knowledge acquisition. The gap between actual outcome and expected outcome that occurred in the transformation process is added to the initial knowledge through a feedback process. The transformation process and the feedback process constitute learning-by-doing.
Finally, every member of the EIP is committed to transfer knowledge to the other members and receive it from others. In general, the ultimate goal of the EIP is to integrate knowledge that was previously distributed to individuals. Only if members of the EIP actively transfer their own knowledge to other members and receive knowledge from others can the EIP become a true knowledge portal instead of a simple Web site that posts information. Knowledge transfer is mostly carried out through communication between members. In fact, the EIP supports direct and indirect communication between its members by using advanced technologies such as electronic mail, groupware, etc. Knowledge transfer is not limited to members who conduct similar tasks within the same division, but it also occurs between members who are charged with different tasks and who learn from each other. In the activity perspective, communities of practice are involved in this type of knowledge acquisition, which is called learning-from-others.
[FIGURE 1 OMITTED]
These three distinct activities, (1) learning-by-investment, (2) learning-by-doing, and (3) learning-from-others, become the major sources of learning where each member of the EIP acquires specialized knowledge.
The outcome of learning processes is measured by the accumulated task-related knowledge and it is evaluated based on two distinct attributes of knowledge: depth and breadth. The depth of knowledge indicates how much knowledge is focused and pertinent in its content. Due to its connection with the specific task, the depth of knowledge is closely related to the level of specialization in tasks, which are assigned to members. Knowledge is also measured in terms of its breadth, which represents the diversity of knowledge across members (Turner et al. 2002). The division of labor affects both depth and breadth of accumulated knowledge. When the division of labor is high, members of that EIP may improve the depth of knowledge through learning-by-doing. However, the high level of division of labor negatively influences the breadth of knowledge, because it limits the diversity of an individual member's task and vision. When it is required to enhance members' knowledge in terms of its breadth, the EIP should expedite different types of learning processes, which are learning-by-investment and learning-from-others between distinct divisions. Figure 1 depicts the three learning processes.
The levels of accumulated knowledge through the three learning processes are influenced by a variety of factors including decay rate of knowledge (Pakes and Schankerman 1979), discount rate of cost (Dorroh et al. 1994), the number of participants joining knowledge creation and transfer activities (De Liso et al. 2001), initial level of knowledge of the member, the amount of knowledge transferred from other members, and the productivity of the learning-by-doing process. We discuss these factors in the research model section.
Background and Research Model
The focus of the current paper is the three learning processes for knowledge accumulation of EIP members and the strategy of knowledge investment and efforts over time. Accordingly, this study presents the optimization problem that maximizes the total profit, which consists of the revenue from accumulated knowledge and costs incurred by the three learning processes. In this section, the effect of the three learning processes on accumulated knowledge and the factors (i.e., discount rate for knowledge and cost, the number of members, etc.) influencing revenue and costs are discussed.
Three Learning Processes
Learning-by-Investment and Learning-by-Doing
Learning-by-investment and learning-by-doing represent the comprehensive knowledge acquisition processes that each member follows while participating in the EIP. Prior research has focused to an extent on learning-by-investment and learning-by-doing. Ba et al. (2001) address an organization's investment in internal knowledge. The objective of their model is to maximize the net surplus resulting from trading knowledge bundles, where knowledge providers offer a bundle of knowledge and auction bids are placed by knowledge consumers who desire it. In other studies in economic literature, learning is considered a production experience measured as accumulated output relative to inputs of labor and investment (Arrow et al. 1961). Dorroh et al. (1986, 1994) develop an economic model explaining the learning process that occurs during production. Their model addresses not only the learning-by-doing effect but also the learning-by-investment effect. They define cumulative knowledge as a direct output resulting from investment as well as a byproduct of production experience. The model developed by De Liso et al. (2001) offers an essential background based upon which we represent the relationship between learning-by-doing processes and division of labor. They assume that the level of the division of labor in the organization grows with more labor input. To determine the influence of division of labor on learning effect, they characterize the Cobb-Douglas production function by formulating the returns to scale of the learning-by-doing effect to be an increasing value as division of labor intensifies.
Following the above literature, we adopt a similar type of Cobb-Douglas production function to represent learning-by-investment and learning-by-doing effects. The fundamental difference between this study and the studies by Dorroh et al. is that the return to scale, which indicates the rate of producing cumulative knowledge, is not merely a given parameter as they assume, but rather is dependent on the level of division of labor in the EIP organization. We apply the logistic function to define the returns to scale of the learning process as a function of division of labor. Another difference between the model in this paper and the models used in prior research is in the measurement of knowledge. De Liso et al. perceive knowledge as a one-dimensional measure and they assume that knowledge simply increases with higher division of labor. The model herein, however, differentiates the influence of the division of labor in terms of two distinct measures of knowledge, depth and breadth, based on the claim that division of labor has different impacts on knowledge depending on its distinct attributes.
The level of division of labor has a significant impact on an individual's knowledge acquisition. When the level of division of labor is high, the task of each individual member becomes more specialized and consequently the members can enhance the depth of knowledge by focusing on the limited area of their specialized tasks. On the other hand, since the focus of each individual member's work becomes narrower within a limited area, members of an EIP having a high level of division of labor may lose chances to improve the breadth of knowledge. Division of labor is positively correlated with the depth of knowledge, but is negatively correlated with the breadth of knowledge. Therefore, this study formulates the return to scale of the learning process in terms of two separate functions that show distinct influences of division of labor on different attributes of knowledge.
Learning-from-Others
In contrast to the previous two types of learning processes, the learning-from-others...
|
