Modeling of tacit knowledge in industry: Simulations on the variables of industrial processes

doi:10.1016/j.eswa.2014.09.023

Expert Systems with Applications

Volume 42, Issue 3, 15 February 2015, Pages 1613-1625

https://doi.org/10.1016/j.eswa.2014.09.023 Get rights and content

Highlights

•
Work objective was to map tacit knowledge into agents of the manufacturing process.
•
It was possible to map the known type to be elicited by the use of Pareto Chart tool.
•
The mapping of critical issues before the decision to elicit tacit knowledge.
•
Bayesian networks based on production rules to predict adverse situations in process.

Abstract

The paper presents the application of a Technical Mapping and tacit knowledge elicitation in industry in order to promote the modeling of tacit knowledge to explicit and represent it in the form of production rules for use in manufacturing processes. The technique was applied with the involved people in the lithographic process in a Metallurgical Company located in southern Brazil. Knowledge of two production coordinators were modeled. For the process of knowledge acquisition and mapping of attributes and values to feed the knowledge base of an expert system, were used quality tools such as Brainstorming, Pareto Chart and Ishikawa Diagram associated with knowledge elicitation techniques such as unstructured interview, rating chips, observation technique, limitation of information and protocol analysis. Quality tools and techniques of knowledge elicitation were systematized to promote process mapping and the elicitation of tacit knowledge, with the aim of representing knowledge by means of production rules. We constructed two knowledge bases with the same methods of production, one in a non-probabilistic expert system (knowledge-based system) and the other in a probabilistic expert system (Bayesian networks) in order to perform comparisons and simulations of the results found. Expert systems perform systematic analysis from the answers given by those involved in lithographic labels process while the defect is identified in order to support the user in diagnosing the root cause of the failure process. From simulations of changes in process variables was possible to prove the hypothesis of the use of probabilistic expert system as industrial support tool in preventing the occurrence of defects in the process and result in a productivity gain.

Introduction

According to Bertalanffy (1973), physiological limitations and paradigms, which determine human capacity, limit the perception of reality in overall phenomena. From this observation, one is faced with the need for employing computing tools so that they can assist in understanding those phenomena. In this study, the phenomenon is applied in such (a) way(s) that the root cause(s) of the defect(s) may be diagnosed during the assemblage of Metal Aerosol Packaging in a Metallurgical Company, its application being intended for industrial processes in general.

The state of the art about how is made knowledge acquisition Wright (1987), Hoffman, 1987, Liou, 1990, Cooke, 1994, Hoffman et al., 1995, Cairo, 1998, Tang et al., 2008, Kim, 2014, Schreiber et al., 2000, Alwis and Hartmann, 2008, Pitchforth and Kerrie, 2013, Oguz and Sengun, 2011, Lemos and Luiz, 2012, Lemos (2012), Kim et al., 2011, Kim, 2014 pursued during the preparation of this study, has shown that there is not a methodology to mapping and elicitation for the of tacit knowledge acquisition in industry, but rather the use of knowledge elicitation techniques used in isolation, without a logical sequence and objective oriented to analysis and problem solving. The methodology has been created based on a systemic approach, which is used in the observation process of complex phenomena in the industry. Concepts of Artificial Intelligence (AI) specifically expert systems (ES), have been used in the research to support the value of tacit knowledge in problem solving inside industries, as a way of disseminating it in internal processes, as well as promoting the organizational learning.

Rezende (2003) says that the main goal of AI is to enable the computer to perform human functions; so, the incorporation of knowledge proves to be essential for the success of any Intelligent System. This statement contemplates the proposed study, which consists in mapping the type of collective tacit knowledge relevant to the solution of the phenomenon observed, thus generating a knowledge management system.

According to Alwis and Hartmann, 2008, Lemos and Luiz, 2012 the transference process of knowledge from tacit to explicit within innovation management in organizations is a competitive advantage once explicit knowledge is already used in the organization; thus, it can be copied by competitors. On the other hand, tacit knowledge is new and, as such, the company will be able to remain for some time with a competitive advantage on its side.

As scientific methodology to structure the sequence of activities used for eliciting tacit knowledge, a technique called systemography has been used. It allows its users to approach, understand and interpret the phenomenon in a systematic way. During the application of the sequence of activities, some quality tools were implemented such as Brainstorming, Pareto Charts and the Ishikawa Diagram. These are tools often used to solve problems in industrial processes.

The objective of this research is to build the methodology, which aims to transform collective tacit knowledge into explicit by using knowledge elicitation techniques, associated to quality tools structured by systemography, represent it in a symbolic language and production rules, and model it in two expert systems which can assist the investigation of defect causes during the metal packaging production process. The production rules, which have been created, were stored in the knowledge’s foundation in two different types of expert systems: probabilistic and non-probabilistic, to be utilized by different users. The first one mentioned will be used at the operational level and the second one at the tactical level of the organization. The expert system tools used were EXPERT SINTA and NETICA. The methodology is named MACTAK – Methodology for Acquisition of Tacit Knowledge.

Section snippets

Theoretical foundation

This section presents issues related to the study’s development, such as: systemography (systemic approach to observation of complex phenomena), quality tools, elicitation knowledge techniques, tacit knowledge, Artificial Intelligence and expert system.

Methodology

The method used in the study was classified as exploratory as its main goal. An exploratory research on the theory forward to the scientific community in the areas of Artificial Intelligence and knowledge management, in order to better adapt the methodology for mapping and elicitation of collective tacit knowledge to the reality is intended to meet Gil (1991) was performed. The methodology, second Santos (2000), can also be classified as action research, because the researcher is one of the

Results and discussion

This section demonstrates the application of the proposed methodology for elicitation of collective tacit knowledge, since the mapping of the type of knowledge to be elicited by the development of production rules, form chosen for formal and explicit representation of tacit knowledge.

Conclusions

The research regarding knowledge acquisition revealed that the practices of knowledge elicitation systematically presented in this work are still little explored in the industries and organizations in general, especially on the optical mapping of the type of knowledge to be elicited. The tacit knowledge is a kind of hard to be voiced to others knowledge, which makes it rarely quoted in an active scientific community in industry in manufacturing operations. Thus, the scientific contributions of

References (56)

S. Aguiar
Integração das Ferramentas da qualidade ao PDCA e ao Programa Seis Sigmas
(2002)
R.S.-de Alwis et al.
The use of tacit knowledge within innovative companies: Knowledge management in innovative enterprises
Journal of Knowledge Management
(2008)
S.Z. Amer et al.
Using group brainstorming in industrial design context: Factors inhibit and exhibit
Procedia – Social and Behavioral Sciences
(2012)
O. Arsene et al.
Medicine expert system dynamic Bayesian network and ontology based
Expert Systems with Applications
(2011)
J.M. Barreto et al.
Análise de falhas no processo logístico devido a falta de um controle de qualidade
(2005)
L.V. Bertalanffy
Teoria Geral dos Sistemas
(1973)
K.K. Bharadwaj et al.
A parallel genetic programming based intelligent miner for discovery of censored production rules with fuzzy hierarchy
Expert Systems with Applications
(2010)
G. Bittencourt
Inteligência Artificial - Ferramentas e Teorias
(2006)
M. Brassard
Qualidade Ferramentas para uma Melhoria Continua
(1985)
M. Brassard
Qualidade: Ferramentas para uma Melhoria Contínua
(1994)

B. Brocka et al.

Gerenciamento da Qualidade

(1994)

O. Cairo

A comprehensive methodology for knowledge acquisition from multiple knowledge sources

Expert Systems with Application

(1998)

Carneiro, P. R., Grimoni, J. A., & Udaeta, M. E. M. (2007). Metodologia e regras para um sistema especialista na...

H. Coelho

Inteligência Artificial em 25 Lições

(1995)

N.J. Cooke

Varieties of knowledge elicitation techniques

International Journal of Human–Computer Studies

(1994)

M. Daychouw

40 Ferramentas e Técnicas de Gerenciamento

(2007)

V. Delcroix et al.

A Bayesian network for recurrent multi-criteria and multi-attribute decision problems: Choosing a manual wheelchair

Expert Systems with Applications

(2013)

A.M.R. Fernandes

Inteligência Artificial: noções gerais

(2005)

Fontana, R. M., & Neto, A. I. (2006). Uma abordagem sistêmica da Mudança Organizacional gerada na implantação de...

A.C. Gil

Como elaborar Projetos de pesquisa

(1991)

L.G.S. Gomes

Reavaliação e Melhoria dos Processos de Beneficiamento de Não Tecidos com Base em Reclamações de Clientes

(2006)

R.R. Hoffman

The problem of extracting the knowledge of experts from the perspective of experimental psychology

Artificial Intelligence

(1987)

R.R. Hoffman et al.

Eliciting knowledge from experts: A methodological analysis

Organizational Behavior and Human Decision Processes

(1995)

J.S. Kim

Development of a composite knowledge manipulation tool: K-expert

Expert Systems with Applications

(2014)

J. Kim et al.

The cognitive selection framework for knowledge acquisition strategies in virtual communities

International Journal of Information Management

(2011)

F.E. Kintschner

(2005)

F.E. Kintschner et al.

Reengenharia de Processos: Transformando as necessidades do cliente em parâmetros de um sistema

(2004)

H. Kume

Métodos Estatísticos para Melhoria da Qualidade

(1993)

Cited by (25)

Systematic review of the “semantic network” definitions
2022, Expert Systems with Applications
Citation Excerpt :
In this category, we discuss the meaning of vertices in a semantic network. For Abdurahiman and Paul (1994); Al-Hassan et al. (2015); Barba et al. (2015); Bingi et al. (1995); Bomberger et al. (2007); Bryniarska (2019); Camargo et al. (2012); Cataldo and Rinaldi (2010); Chen et al. (2020); Cheng et al. (2019); Frederiksen (1975a); Frederiksen (1975b); Ge et al. (2022); Hersh and Greenes (1990); Higuchi et al. (1991); Jauregui-Becker et al. (2009); Kareem and Allen (1990); Klusch (1994); Li (2022); Liang et al. (2021); Lim (2019); Maceratini et al. (1989); Marinov and Zheliazkova (2005); Matsatsinis et al. (1997); Moldovan and Tung (1985); Nault and Storey (1998); Partridge (1996); Rasheed et al. (2017); Rinaldi (2014); Robinson (1990); Rosario et al. (2015); Shapiro (1977); Shastri (1989); Shoval (1985); Simeone et al. (2019); Singleton (1991); Sowa (1991); Stantchev et al. (2015); Suh et al. (1995); Wolfengagen et al. (2021); and Wu and Wysk (1990), vertices represented concepts. Vertices could also represent objects in the definition by the following: Abdurahiman and Paul (1994); Bingi et al. (1995); Kareem and Allen (1990); Maceratini et al. (1989); Robinson (1990); and Stantchev et al. (2015); as well as data items in the definition by Moldovan and Tung (1985); events in the definition by Chen et al. (2020) and Maceratini et al. (1989); situations in the definition by Abdurahiman and Paul (1994), Bingi et al. (1995), and Rosario et al. (2015); facts in the definition by Shoval (1985) and Suh et al. (1995); pieces of data in the definition by Shoval (1985); semantic symbols in the definition by Frederiksen (1975a); objectives in the definition by Rosario et al. (2015); and instances in the definition by Cheng et al. (2019).
Knowledge can be represented through semantic networks, which have the advantages of being simple, clear, and easy to visualize. The literature points to the existence of a diversity of “semantic network” definitions, which can be conceived of as structures based on graphs with edges or arcs. According to this assumption, vertices can be assigned meanings or be labeled, and they can represent concepts, objects, or individuals; edges can have direction or be labeled, and they represent relationships or associations. A semantic network model can be designed to capture only one or several aspects of a semantic structure. We performed a systematic review of the literature with the aim of identifying and classifying the family of “semantic network” definitions adopted by the authors of the studies reviewed. This study mapped the “semantic network” definitions present in the literature and analyzed aspects such as the mathematical representation used, vertices and edges representation, the attribution of meanings to each network component and the semantic network representation. We also analyzed the core of words linked to the family of the “semantic network” term in a semantic network based on their definitions.
Using Bayesian belief networks to improve distributed situation awareness in shift changeovers: A case study
2022, Expert Systems with Applications
This paper presents a case study about communication between workers during shift changes based on empirical research at a continuously operating industrial operation.
The premises of Situation Awareness (SA) and Distributed Situation Awareness (DSA) are the theoretical lens for conducting the present study. The objective of the study is the probabilistic modeling of the SA of process operators by means of complementarity among the premises of SA and DSA during shift change.
The methodology used semi-structured interviews that followed a questionnaire created to categorize verbalizations at the three SA levels. Verbalizations were analyzed with the “digital humanities” approach that allowed the integrated modeling of SA using Bayesian Belief Networks (BBN) theory. The research was performed in an industrial plant that carried out a polymer injection process, which was classified as a physical arrangement for the continuous manufacture of products. In this process, production was conducted in three continuous shifts with one operator per shift.
As a result, the BBN modeling showed the dynamic relationships among the core variables monitored, and also demonstrated the complementarity between SA and DSA. In this sense, BBN was used as a communication protocol between shifts and served as a mechanism that influences the operator’s cognitive load. The developed algorithm showed an excellent performance, because it represents the conditional probability between the variables according to the operators' indications regarding the criticality and level of monitoring of the process. The modeling allowed representing the operators’ complete SA, thus promoting DSA.
Expert Systems in Special Machinery: Increasing the Productivity of Processes in Commissioning
2017, Procedia CIRP
Due to the megatrend globalization, special machinery is gaining significance for the capital goods sector. Characterized by the fulfillment of individual customer requirements, companies in special machinery have to deal with very specific and technologically complex tasks. Hence, managing information and knowledge becomes vital for a company's competitive ability, notably when it comes to expert knowledge. The characteristics of special machines leads to iterative processes for problem solving and thereby, increase lead times significantly. The more technologically complex a machine is, the more scattered the expert knowledge, meaning that many different experts need to be consulted before solving a problem. Up to now, in scientific literature, there has been little discussion about the challenges of special machinery and practical solutions regarding an implementation of technical intelligence in a special machinery environment. Therefore, the goal of this paper is to give an example of how an expert system can be applied to special machinery surroundings and thus, increases productivity. A Bayesian network forms the basis of the system as it allows efficient inference algorithms and reasoning under uncertainty, despite its ability to describe complex dependencies. The expert systems capability has been proven in industrial laser manufacturing.
A framework to acquire explicit knowledge stored on different versions of software
2016, Information and Software Technology
Citation Excerpt :
As a result studies that could be applied to re-acquire explicit knowledge as well as some methodologies where found, however, most of them focus on tacit knowledge; the distinction between these two types of knowledge has been widely studied [12,4], a possible reason for it is the perception that tacit knowledge is more difficult to share and not readily codified [16] sometimes the production of new knowledge has more weight than the exploit of the one that is already captured [32,33]. The conversion to explicit knowledge to be used on KM tools [34], initiatives to manage knowledge through including database integration and conversion, internet architectures, artificial intelligence or technical support for decision making [35] corporate memory generation focused on efficient representation of tacit knowledge [23], development of frameworks to facilitate transfer [33] or how to get interaction between groups [36] are some study cases on acquisition and re-use of knowledge. After the review, the lack of studies on re-acquisition of explicit knowledge is evident, most of the studies and frameworks assume that if knowledge is already encoded and captured it can be exploited.
In organizations, especially small and medium ones (SME's), explicit knowledge is stored in different sources of hardware or software. These sources have become obsolete due to constant technological innovation and the growing obsolescence of hardware, and have made the knowledge they contain unreachable. These organizations are forced to seek alternatives that allow them to use the knowledge they already possess, but are unable to exploit.
To present and execute a framework for the acquisition, classification and dissemination of explicit knowledge among the organization for those who may use that knowledge in their daily activities.
A framework is developed to acquire the explicit knowledge from different sources and versions of software. Its efficiency has been verified through its' application in a pharmaceutical company.
The knowledge has been transferred from an unreachable place so that it can be exploited by the organization. Access control has been added for increasing the level of security where the knowledge is stored. In the implementation, the framework was tested with a general application presenting very favorable results.
In general, the effectiveness of the framework was tested, showing that knowledge recovery from different versions of software is possible. In that way, the problem observed on SME's can be solved in an efficient and replicable form.
Thread-based probabilistic models for expert finding in enterprise Microblogs
2016, Expert Systems with Applications
Citation Excerpt :
An ‘expert’ is typically defined as a person with relevant ‘expertise or experience’ for a given query; and an ‘expert-finding system’ is a type of intelligent search engine that retrieves ‘a list of ranked expert candidates who match the expertise query’ (Becerra-Fernandez, 2006; Ehrlich & Sadat Shami, 2008). In the research reported here we thus contribute to the following aspects of expert and intelligent systems (Liao, 2005; Sahin, Tolun, & Hassanpour, 2012; do Rosário, Kipper, Frozza, & Mariani 2015; Vásquez-Bravo, Sánchez-Segura, Medina-Domínguez, & Amescua 2014): Establish that business conversation shifting to Enterprise Social Media provide a unique source of tacit knowledge that merits further research into useful knowledge acquisition methods. (
Within the enterprise important conversations are shifting to Social Media Microblogs thus providing a new source for the acquisition of tacit knowledge useful for intelligent agents. While automated retrieval of candidate experts from Web documents has been well-researched, little exists that leverages social aspects of Microblogs for this purpose. We show that the experts and their related expertise can also be identified in the enterprise corpus. Analysis reveals the existence of problem-solving threads where the requestor seeking help often elicits responses from experts which consequently records tacit (experiential) knowledge. Here we present rules to automate the acquisition of this tacit knowledge. The rules are based on probabilistic models enhanced with linguistics to exploit the role patterns in threads. Heuristics also strengthen local evidence about associations between candidates, documents and (expertise) terms. Applying this, we demonstrate that the Enhanced Models can ameliorate the negative impact of Microblogs (such as sparse data, short content, and implicit associations). The experiments show that the candidate models significantly outperform the document models in the Microblog environment. This is different from previous research in Web environments. Examples also illustrate underlying insights and emphasize the ‘additive’ nature of expertise found in Social Media.
On tacit knowledge management in product design: status, challenges, and trends
2024, Journal of Engineering Design

View all citing articles on Scopus

View full text

ReviewModeling of tacit knowledge in industry: Simulations on the variables of industrial processes

Highlights

Abstract

Introduction

Section snippets

Theoretical foundation

Methodology

Results and discussion

Conclusions

Integração das Ferramentas da qualidade ao PDCA e ao Programa Seis Sigmas

The use of tacit knowledge within innovative companies: Knowledge management in innovative enterprises

Journal of Knowledge Management

Using group brainstorming in industrial design context: Factors inhibit and exhibit

Procedia – Social and Behavioral Sciences

Medicine expert system dynamic Bayesian network and ontology based

Expert Systems with Applications

Análise de falhas no processo logístico devido a falta de um controle de qualidade

Teoria Geral dos Sistemas

A parallel genetic programming based intelligent miner for discovery of censored production rules with fuzzy hierarchy

Expert Systems with Applications

Inteligência Artificial - Ferramentas e Teorias

Qualidade Ferramentas para uma Melhoria Continua

Qualidade: Ferramentas para uma Melhoria Contínua

Gerenciamento da Qualidade

A comprehensive methodology for knowledge acquisition from multiple knowledge sources

Expert Systems with Application

Inteligência Artificial em 25 Lições

Varieties of knowledge elicitation techniques

International Journal of Human–Computer Studies

40 Ferramentas e Técnicas de Gerenciamento

A Bayesian network for recurrent multi-criteria and multi-attribute decision problems: Choosing a manual wheelchair

Expert Systems with Applications

Inteligência Artificial: noções gerais

Como elaborar Projetos de pesquisa

Reavaliação e Melhoria dos Processos de Beneficiamento de Não Tecidos com Base em Reclamações de Clientes

The problem of extracting the knowledge of experts from the perspective of experimental psychology

Artificial Intelligence

Eliciting knowledge from experts: A methodological analysis

Organizational Behavior and Human Decision Processes

Development of a composite knowledge manipulation tool: K-expert

Expert Systems with Applications

The cognitive selection framework for knowledge acquisition strategies in virtual communities

International Journal of Information Management

Reengenharia de Processos: Transformando as necessidades do cliente em parâmetros de um sistema

Métodos Estatísticos para Melhoria da Qualidade

Review
Modeling of tacit knowledge in industry: Simulations on the variables of industrial processes