Vol.3 No.1 2010

83/110

Research paper : Acquisition of skills on the shop-floor (N. Matsuki)−80−Synthesiology - English edition Vol.3 No.1 (2010) 4 Research resultIn this chapter, the skill acquisition method that was finally organized will be described. More accurately, the framework for acquisition is presented. This framework fails to propose the meta-structure for skill acquisition that allows acquisition of a desired skill with ease. However, it may provide a certain viewpoint when acquiring the skills and when considering the meaning of the expert skilled worker in future manufacturing.4.1 Overall structureThe result of this research can be organized as shown in Fig. 2. The computer system as an alternative has a derived model for outputting the decision value. The derived model is composed of the derived algorithm that calculates the decision value by entering measurements or some derivable values from the work environment.The excellence of an expert skilled worker that became clear by studying this derived model is the “ability to simplify the problem.” For example, in forging, the important role of the expert skilled worker is to calculate the pressure needed for manufacturing the product. Here, the expert worker seems to simplify and categorize the complex product shape to calculate the processing pressure. It is assumed that by doing so, the parameters to be considered can be greatly reduced and the sufficiently precise answer can be given for deciding whether a product can be forged using the machines at the company. The derived model as an alternative is quite different from an expert worker, yet there is a similarity that the complex phenomenon is simplified. As the research progressed, we thought that the greatest characteristic of an expert skilled worker was the ability to simplify the subject and to quickly come up with reasonably accurate answers. The ability to discern the simplification process is the true value of the expert skilled worker.Simplification is important in the derived model for alternatives. The on-site environment is often complex, and it often becomes an extremely complex and unstable system if all the influences of the related factors are taken into account. It is necessary to build the derived model to enable decision-making by selecting the truly dominant factors. Also, obtaining useful results from simple input means that the operation on site is easy. The limitation of “developing a technology that can be used on site” actually played an important role in discovering the ways of simplification. In this research, the condition that it must be usable on site was effective in keeping the research activity sound.Looking at the skill transfer tools created, they can be organized into several types. Following are the explanations of the types.4.2 Derivation by theoretical formulaIn the cold forging treatment, the desired part is shaped by stamping the metal at high pressure. It is known that the deformation resistance and restraint coefficient play important roles as shown in Fig. 3 in the process of metal deformation. From various experiences, it could be estimated that approximate value can be obtained by fitting the shapes of the part shape into some patterns for the purpose of considering the forging pressure, even though the shape of the parts are different. It was also found that by using iron as a standard for estimating the property values, this could be applied to other materials such as magnesium and aluminum.The derived model for pressure values needed in cold forging was created from the companies’ experience and knowledge of metal plasticity, as well as the drawings and materials of the parts to be manufactured. Figure 4 shows the interface of the implemented computer system. We were also able to create a derived model for the temperature increase of the product during forging.Fig. 2 Structure diagram of the counterpart to decision-making skills. Derived modelDerived valueEnvironment(parameter)Derived algorithmAcquisition ofinput parameterDeformation resistance ofcarbon steelNo frictionNo lubricationUpper limit of restraint coefficientLower limit of restraint coefficientRestraint coefficient ofaxisymmetric shapeProcessing pressure = Restraint coefficient × Deformation resistance①②③④⑤①:S10C(0.09 %C, 105 HV)②:S20C(0.21 %C, 120 HV)③:S30C(0.32 %C, 143 HV)④:S40C(0.40 %C, 170 HV)⑤:S50C(0.53 %C, 185 HV)250020001500100050000204060801001234501234567Restraint coefficient p/YDeformation resistance Y, MPaLogarithmic strain ln (h0/h)Percentage of reduction r, %Backward extrusion process of containerForward extrusion process of shaftFig. 3 Model for calculating the processing pressure in forging.

※このページを正しく表示するにはFlashPlayer9以上が必要です