IMPORTANT IDEAS ARE OFTEN surprisingly difficult to define. Most theologians, for example, would struggle to adequately define “God,” just as a professor in political science would be hard-pressed to explain in few words what exactly constitutes “politics.” Among emergent business concepts, few are more important than knowledge management (KM). Due to the increasing emphasis on the value of knowledge work and information technology in the global economy, KM is a critical component of any successful business strategy (Laudon & Laudon, 2002) Despite a certain “faddishness,” as one article puts it, evident in the many conferences, workshops, books, and tapes dealing with the subject, KM is also here to stay: “There are far too many knowledge workers dealing with too much knowledge for knowledge management to disappear” (Davenport & Grover, 2001, ¶ 1). As popular as the concept is, however, precise and widely accepted definitions of KM are hard to come by even among experts in the field (Wainwright, 2001; Horwitch & Armacost, 2002) – a fact which perhaps further suggests the importance of this fairly novel management discipline within the larger field of management information systems (MIS). This paper is an attempt to provide a sociotechnical understanding of KM by exploring related concepts – artificial intelligence and emotional intelligence – representing both the technical and behavioral aspects of KM and information systems management.
An Interdisciplinary Approach to Knowledge Management
Laudon and Laudon (2003) define “knowledge management” in terms of processes “developed in an organization to create, store, disseminate, and apply the firm’s knowledge” (p. 317). Similarly, KM has been defined elsewhere as “a process that helps organizations find, select, organize, disseminate, and transfer important information and expertise” to facilitate other important business activities in turn (Gupta, Iyer & Aronson, 2000, ¶ 1). Observing the failures of various KM programs in the past, some researchers are careful to assess KM primarily in terms of its perceived practical usefulness as a set of business processes (Horwitch & Armacost, 2002). Others reach a similarly pragmatic conclusion, noting that because KM is simply beyond strict academic definitions-since “knowledge” itself is ultimately a philosophical notion-it should be carefully monitored on the basis of enhancing organizational performance and providing direction or justifiable grounds for action (Wainwright, 2001; Santosus & Surmacz, 2001).
This lack of precise definition is likely due to the fact that KM, like the sociotechnical field of MIS in general, incorporates divergent technical and behavioral disciplines (e.g., computer science, information technology, psychology and economics). It follows that KM cannot be definitionally reduced to the mere use of information technology, though it often is defined this way nonetheless (Santosus & Surmacz, 2001). Wide ranging interdisciplinary concepts are naturally more difficult to define than more isolated concepts. Therefore an examination of KM with the use of specific examples from both a technical field (computer science) and a behavioral field (psychology) should prove illuminating-all the more so since the two specific subjects to be examined, artificial intelligence and emotional intelligence, have both emerged as popular and promising areas of interest for business in general.
Artificial Intelligence and Explicit Knowledge
Most of us grew up in what could be termed an industrial society, in which mass production and blue-collar labor dominated business and commerce. Much evidence indicates that the blue-collar industrial economy is quickly becoming a white-collar information economy (Laudon & Laudon, 2002; Wainwright, 2001). In this increasingly globalized information economy, knowledge (along with the capacity to quickly obtain and utilize it) has a definite cash value (Wainwright, 2001). Information technology (IT) undoubtedly figures prominently in the growing importance of knowledge, the training of knowledge workers, and the development of knowledge work systems in the information economy. Indeed, Laudon and Laudon (2002) list office systems, knowledge work systems, group collaboration systems and artificial intelligence (AI) applications among the most vital technologies available to facilitate KM. In particular, AI represents one of the more technical and potentially valuable approaches to the management of knowledge-especially explicit knowledge, or in other words “more objective, rational, and technical knowledge” (Gupta, et al, 2000, ¶ 4). Some would contend that AI also plays a role in the capture and manipulation of tacit knowledge, but the point here is that AI technology itself is essentially a means of processing data, which amounts to explicit knowledge.
Unfortunately, frightful and fanciful mischaracterizations on the part of the popular media and some now disappointed theorists have given AI a bit of a bad name, so that it briefly went out of style during the late eighties and most of the nineties (Buck, 1997; Derra, 1999). However, whereas AI robots have failed to take over planet Earth, a kindler, gentler sort of AI has in the meantime found a place in the world of practical business applications-in the form of distinct advanced technology tools used to aid in decision making and the capture of knowledge (Buck, 1997; Derra, 1999). Rather than trying to design a superintelligent machine that somehow passes for a human by passing the “Turing Test,” many researchers are now discovering more profitable (if more modest) uses for specific AI tools – e.g., expert systems, neural networks, fuzzy logic and genetic algorithms. The emerging focus in AI research therefore involves its practical utility in real-world applications, “acting as a filter to help humans make wise decisions in a world awash in information” (Derra, 1999, ¶ 5).
Businesses today may have in place any number of specific AI applications that capture and use explicit knowledge in order to optimize information storage, provide an objective (unemotional) means of decision making, and update the business’ knowledge base by rapidly generating solutions to complex problems beyond the scope of individual analysis (Laudon & Laudon, 2002). Now widely used in business, expert systems for example simulate human reasoning by the use of inference engines and a knowledge base pertaining to a particular area of expertise. Expert system applications include investment analysis, management and forecasting; enhanced spreadsheet and report generation offering advice to managers; and diagnostic tools used in fields as far ranging as medicine and auto repair (Qureshi, Shim & Siegel, 1998). The medical profession especially has benefited from expert systems, using them to accurately and efficiently diagnose specific diseases, suggest the presence of combinations of illnesses, and even assess - with 95% accuracy - death prognoses (Metaxiotis & Samouilidis, 2000).
Another form of AI technology, neural networks are essentially computer-based mathematical simulations of the human decision making process, at least to the extent that it can be physically observed within the nervous system (Qureshi, et al, 1998). Unlike expert systems, neural networks do not reach deductive solutions to specific problems, but rather analyze large quantities of data to recognize patterns, make predictions or otherwise make what amount to inferential best guesses (Laudon & Laudon, 2002). Neural nets, and the nonlinear “fuzzy logic” approach associated with them, consequently require less formal program coding but much more data and hence more sheer processing capability. Various organizations are beginning to discover specialized applications for neural net technology: Analyzing pap smears for abnormalities; predicting financial equity performance and corporate bankruptcies; and searching credit card transaction records for changes in buying patterns in order to detect fraud (Laudon & Laudon, 2002). North Star Steel has employed “optimizing” neural systems to design and monitor more efficient furnaces, and Symantec’s antivirus software features neural technology capable of detecting computer viruses much earlier than was previously possible (Derra, 1999).
Likewise rapidly increasing in popularity are hybrid systems, or combinations of AI technologies into a single business application (Laudon & Laudon, 2002). CAD/Cam is a product developed by AI Ware that integrates neural networks, genetic algorithms and sensitivity analyses to “discover the underlying effects between a number of process parameters” in the creation of pharmaceutical, chemical and petroleum products (Studt, 1997, ¶ 12). Hybrid techniques have also proven fruitful in the field of financial engineering (including corporate finance, investment, portfolio management, etc.), outperforming not only conventional predictive and analytical methods, but also stand-alone AI tools such as expert systems (Monostori, 1999). Along with knowledge work systems and group collaboration systems based on Internet technology, AI research will continue to produce technology tools that aid in the management of explicit knowledge for the emerging “knowledge economy” (Wainwright, 2001).
Emotional Intelligence and Tacit Knowledge
Knowledge management has to do with organizational learning-the process of using acquired knowledge to enhance business processes-which in turn requires (among other things) continual feedback from people (Laudon & Laudon, 2002). Organizational learning therefore depends on informal, tacit knowledge. As opposed to explicit knowledge, tacit knowledge is information “in the raw,” based on fairly subjective human insights, interactions, intuitions and experiences (Gupta, et al, 2000). Because all explicit knowledge is really so much tacit knowledge “converted” into codified form, it follows that tacit knowledge is essential to any knowledge-based undertaking. Despite these facts, there exists a tendency among technical KM theorists to devalue tacit knowledge and the human element that is its source: “The knowledge economy appears not to value this form of knowledge” (Tomlinson, 2002, ¶ 13).
Moreover, there are indications that in many areas of business, information technology has been misapplied and is therefore actively counterproductive. Witschger (2000) cites the example of the “automated attendant,” an extension of the voice mail concept designed to replace the traditional receptionist. Built on the premise that technology should always be applied whenever and wherever possible, the automated receptionist system has proven frustrating to the businesses using it as well as the callers trying in vain to contact them directly by phone. The reasons for the frustration have to do with the interpersonal nature of a phone conversation and the attempt to blithely replace it with programmed scripts. Witschger (2000) goes on to argue that there is an important distinction to be made between transacting business and conducting business: “You cannot conduct business with a computer, because a computer is not a legal entity nor is it a living entity” (¶ 6).
One particular area of management-related research promises to counteract these dehumanizing trends: Emotional intelligence (EI) is an important concept having to do with informal management techniques and is arguably essential to the success of any KM program. According to EI researchers-including Daniel Goleman, author of Emotional Intelligence and the first to popularize the notion-there are five elements that make up EI (and at the same time serve as distinguishing characteristics between human beings and the technologies they create): (1) self-awareness, especially in terms of emotion; (2) self-regulation, or the ability to control emotions and recover from setbacks; (3) motivation, reflected in commitment, drive and a reasonable degree of optimism; (4) empathy, or understanding, developing and respecting others; and (5) social skills, notably communication, leadership, team-building, and handling of relationships (O’Shaughnessy, 1999; O’Neil, 1996). In an interview with John O’Neil, Goleman defined EI succinctly as “a different way of being smart. It includes knowing what your feelings are and using your feelings to make good decisions in life” (O’Neil, 1996, ¶ 1).
Though specific, quantifiable evidence for it has yet to be outlined in detail (for the moment), a general correlation between EI and the perceived effectiveness of KM in knowledge-based enterprises such as educational institutions and technological networks appears to square well with observations (Tomlinson, 2002; Palmer & Richards, 1999). If the capture of tacit knowledge begins with interpersonal communication (as it seems), then such a correlation is probably sound and can be further applied to KM in general. Some of the most useful knowledge is of the tacit variety, disclosed by people in the know as to what is really happening in the organization. In their definition of knowledge in a KM context, Davenport and Grover (2001) implicitly allude to the value of EI in the culling of tacit knowledge: “Technology can provide assistance in knowledge management, but its importance pales in comparison to developing knowledge-oriented cultures, motivating individuals to share and use knowledge, and encouraging workers to view their jobs in terms of effective knowledge management” (¶ 4). Among KM practices commonly used in the “conversion” of tacit knowledge into explicit knowledge, first on the list is a skill closely related to EI: “Socialization: sharing of experiences through observation, imitation and practice” (Gupta, et al, 2000, ¶ 13).
This “conversion” from tacit, human-based knowledge to explicit knowledge residing on hard drives, databases and networks has everything to do with another conversion-the conversion of the personality due to the effectual communication of leaders (Wieand, 2002). According to Wieand (2002), studying the works of Peter Drucker, effective organizational communication presupposes not only technical competence but a high level of EI and a corresponding sense of trustworthiness, sincerity or authenticity on the part of managers or leaders. The idea is that while technology has changed and is constantly changing still, human nature remains the same (Wieand, 2002). EI is therefore an area of management competency necessary for driving the human side of the organization, by respectfully appealing to the emotions, values and intellectual capacities of people within it in order to receive their best efforts and inputs (Wieand, 2002).
Informal, tacit knowledge ultimately resides in the minds of individuals in the organization (Laudon & Laudon, 2002). Communication, a primary expression of EI, is the means by which these thoughts emerge and are shared with others in the organization. The resulting irony is hard
to miss: The management of knowledge, even of the most scientific, logical, explicit sort, ultimately begins with everyday communication and interpersonal skills: “The most powerful communication may be nothing more, but nothing less than, shared experience, without any logic whatever” (Wieand, 2002, ¶ 8). This is the very sort of experience that has actually defined human relationships for thousands of years: “In some ways, emotional intelligence is really not new” (Cherniss, 2000, p. 6) Numerous strands of evidence nonetheless suggest that EQ (emotional quotient) among managers is a much greater factor than IQ in the determination of everything from overall job performance to sales volume and net profits (Cherniss, 2000). In the growing information economy, the need to develop EI among managers and workers will likely only increase as technological advances and changes continue to challenge the intellectual and psychological capacities of the average worker to absorb them.
Balancing Technical and Behavioral Aspects of Knowledge Management
None of this is to say that information systems or technologies are irrelevant to the success of a business organization. To the contrary, all parties acknowledge that IT advances will continue to proliferate and to provide ample opportunities for businesses to exploit. Indeed, those businesses that fail to track the information explosion and properly manage knowledge assets in their respective fields are almost certainly headed for failure (Laudon & Laudon, 2002; Horwitch & Armacost, 2002). On the other hand, a general failure on the part of managers, analysts and technicians to appreciate the complexities of human behavior in organizations has led to wholesale information systems dysfunction: “Historically, information system design has been preoccupied with technical issues at the expense of organizational concerns. The result has often been information systems that are technically excellent but incompatible with their organization’s structure, culture, and goals” (Laudon & Laudon, 2002, p. 429).
Consequently, the current situation calls for a balanced sociotechnical approach to KM and information systems management in general. This means in practical terms that with every system upgrade or eye-popping technological innovation introduced to an organization, there must be a corresponding effort to familiarize workers with the changes, all the while assuring them that people are in fact “the ultimate technology” (Witschger, 2000). Specific courses of action during a system overhaul, for example, would include training project managers in the use of internal integration tools; conducting an organizational impact analysis; and involving end-users in the establishment of a sociotechnical design (Laudon & Laudon, 2002). Otherwise the new technology gathers dust owing to disuse and mismanagement, while the employees become mystified or even threatened by the machines themselves-some of them going by the name of “artificial intelligence”-for precisely the same reason.
According to Palmer and Richards (1999), these sorts of concerns are legitimate: As the new economy erodes traditional organizational structures and procedures, the result is not (as might be expected) enhanced connectivity but social isolation: a person detached from the world, sitting at a terminal, tapping on a keyboard, staring at a phosphorous screen. A proposed solution is NQ (“network intelligence,” related to and borrowed from Goleman’s EQ concept), which addresses the phenomenon of “social fragmentation that is inherent in technological networks” (Palmer & Richards, 1999, ¶ 55). The KM specialist must therefore actively encourage the formation of a group identity rather than a mere individualized identity, by means of NQ methods of promoting learning, dialogue, sensitivity, and emotionally meaningful interaction within a technological setting (Palmer & Richards, 1999).
Perhaps the lack of a clear definition of KM, as mentioned earlier, has something to do with the problem. “Knowledge management” bears a certain reexamination. In what may prove a serious understatement, Santosus & Surmacz (2001) mention in passing: “It’s important to note that the definition [of KM] says nothing about technology” (¶ 1). That is, while KM and technology are closely related, they are never quite synonymous. Popular notions of KM often tend drastically toward the technical and subtly disregard the human and behavioral. Corporate IT gurus would literally profit, therefore, from embracing the Japanese understanding of KM, which includes not only information processing but means of accessing tacit knowledge and incorporating individual experience and subjective insights (Tomlinson, 2002; Gupta, et al, 2000). An expanded presence of information technology associated with KM-in the form of advanced knowledge work systems and AI tools-actually necessitates more, not less, emphasis on EI, organizational behavior and other distinctly humanistic areas of research. Maintaining a deliberate sociotechnical tension will do much to restore confidence among workers and at the same time help to develop a useful, meaningful and profitable store of knowledge - rather than merely a vast collection of data.
An Illustration of the Right Balance
A closing story taken from the Wall Street Journal about the sharing of information at Buckman Labs illustrates the benefits of a healthy marriage between technology and humanity in the knowledge enterprise: When the people at the new recycling mill at Manistique Papers Inc. realized that the peroxide they used to remove ink from old magazines wasn’t doing the job, they contacted Buckman Laboratories, their chemical supplier (Thurm, 1999). David Crownhart, a sales manager at Buckman, posted a message on the company’s online discussion group message board, seeking answers. Some knowledgeable salespeople from offices in Europe suggested that the problem had to do with enzyme-producing bacteria that were breaking down peroxide in the paper slurry. Crownhart suggested a chemical to counteract the problem, and Manistique reported positive results in their paper turning the desired shade of white (Thurm, 1999).
According to chief executive Robert Buckman, the incident provides a valuable lesson in the sociotechnical realities of KM. Whereas the prevailing theory is that instant communications and superior computing power automatically create a competitive edge, those technologies are only as effective as the people who utilize them (Thurm, 1999). Buckman’s knowledge-rich information system is practically useful for situations such as described above largely because of a serious and protracted effort on the part of Buckman himself to encourage active participation, i.e., the construction of a useful knowledge base. Mr. Buckman is after not only explicit knowledge, but tacit knowledge, what he refers to as “the real gold in the organization” (Thurm, 1999, ¶ 17). Nowadays reading and contributing to the company message board is a required job function for salespeople, and overseeing the 54 individual online forums-while monitoring them for a properly balanced sociotechnical content-is one of the functions of management. As indicated by Thurm (1999), experts are quick to point out the necessity of leading-edge technology in the acquisition and management of knowledge: “But getting people to open up is not as simple as installing a software program” (¶ 7).
References
Buck, Neena. (1997). Just don’t call it AI. Computerworld, 31, 79-80. Retrieved February 18,
2003 from Proquest database.
Cherniss, Cary. (2000). Emotional Intelligence: What it is and Why it Matters [Electronic
version]. Retrieved February 28, 2003 from http://www.eiconsortium.org.
Davenport, Thomas H., & Grover, Varun. (2001). Special Issue: Knowledge Management
Journal of Management Information Systems, 18, 3-4. Retrieved March 21, 2003 from
http://jmis.bentley.edu/articles/v18_n1_p3/index.html
Derra, Skip. (1999). Refocusing AI on real-world applications. Research & Development, 41,
46-48. Retrieved February 20, 2003 from Proquest database.
Gupta, Babita, Iyer, Lakshmi S., & Aronson, Jay E. (2000). Knowledge management: practices
and challenges. Industrial Management + Data Systems, 100, 17-21. Retrieved February
28, 2003 from Proquest database.
Grossman, Robert J. (2000). Emotions at work. Health Forum Journal, 43, 18-22, Retrieved
February 27, 2003 from Proquest database.
Horwitch, Mark, & Armacost, Robert. (2002). Helping knowledge management be all it can be.
The Journal of Business Strategy, 23, 26-31. Retrieved February 27, 2003 from Proquest database.
Laudon, Kenneth C., & Laudon, Jane P. (2003). Essentials of Management Information Systems:
Managing the Digital Firm (5th ed). Upper Saddle River, New Jersey: Prentice-Hall.
Metaxiosis, K.S., & Samouilidis, J.E. (2000). Expert systems in medicine: academic illusion or
real power? Information Management & Computer Security, 8, 75-79. Retrieved February
27, 2003 from Proquest database.
Monostori, Laszlo. (1999). Subsymbolic and Hybrid Artificial Intelligence Techniques in
Financial Engineering. Retrieved March 21, 2003 from http://www.ercim.org/publication/Ercim_News/enw38/monostori2.html.
O’Neil, Paul. (1996). On Emotional Intelligence: A Conversation with Daniel Goleman.
Retrieved March 25, 2003 from http://www.ascd.org/readingroom/edlead/9609/oneil.html.
O’Shaughnessy, John. (1999). Working with emotional intelligence. Journal of Macromarketing,
19, 178-179. Retrieved February 28, 2003 from Proquest database.
Palmer, Joy, & Richards, Ian. (1999). Get knetted: network behaviour in the new economy.
Journal of Knowledge Management, 3, 3. Retrieved February 28, 2003 from Proquest
database.
Qureshi, Anique A., Shim, Jae K., & Siegel, Joel G. (1998). Artificial intelligence in accounting
& business. The National Public Accountant, 43, 13-18. Retrieved February 28, 2003 from Proquest database.
Santosus, Megan, & Surmacz, Jon. (2001). The ABCs of Knowledge Management Retrieved
March 21, 2003 from http://www.cio.com/research/knowledge/edit/kmabcs.html.
Studt, Tim. (1997). Embedding and hybrids key to AI growth. Research & Development, 39,
73-74. Retrieved February 18, 2003 from Proquest database.
Tomlinson, Harry. (2002). Effective Schools for the Future Retrieved March 9, 2003 from
http://www.cybertext.net.au/tct2002/disc_papers/organisation/tomlinson.htm.
Thurm, Scott. (1999). Technology (a special report)-balancing act-what do you know?
Getting employees to share their knowledge isn’t as simple as installing new software; Just ask Buckman Labs. Wall Street Journal, Eastern Edition, R10. Retrieved February 28, 2003 from Proquest database.
Wainwright, Christopher. (2001). Knowledge management: Aspects of knowledge. Management
Services, 45, 16-19. Retrieved February 18, 2003 from Proquest database.
Wieand, Paul. (2002). Drucker’s challenge: Communication and the emotional glass ceiling.
Ivey Business Journal, 66, 32-37. Retrieved January 20, 2003 from Proquest database.
Witschger, Jim. (2000). People the ultimate technology. Manage, 52, 28-29. Retrieved February
27, 2003 from Proquest database.