Vol.2 No.1 2009

32/88

Research paper : Modeling the social acceptance of industrial technologies (M. Matsumoto et al.)−29−Synthesiology - English edition Vol.2 No.1 (2009) coefficient of preference weight wij was set based on the result of conjoint analysis. Level skj of product attribute was set. Also, learning curve model was set to set the price reduction transition of eco-product in the basic case. The change of level of attribute in basic case was set. Other price change factors (such as subsidy), if any, were set. Based on the above settings combined with basic setting, the diffusion was simulated by equation (4). This was the diffusion projection of basic case.Next, the scenario case was set to conduct the scenario analysis. Change of levels of attributes that were different from basic case was set. For example, future changes of attribute levels due to technological advances or government policies such as subsidy and carbon tax were set as scenario. As in basic case, diffusion was simulated by equation (4).5 Data collection5.1 Diffusion coefficient of past productsTo refer to the diffusion coefficients of past products, the diffusion curves of past products were collected and their diffusion coefficients were extracted. Figure 5 shows the diffusion transition of washing machines and the approximated curve by Bass model. In this research, diffusion curves of 28 products were collected. Table 2 shows the diffusion coefficients of 20 products. As guideline of speed of diffusion, the rightmost column of the table shows the years required for the diffusion rate to increase from 10 % to 50 %. It generally took four to five times as long as the years specified in the rightmost column from launch of product to saturation in market. For automobiles, it was 40 to 50 years, while it was about 10 years for home appliances such as television that diffused very fast in the 1960s.5.2 Conjoint analysisIn this research, three conjoint analyses were conducted. They were conducted for 1) home electric appliances about 100,000 yen, 2) home installations about 1 million yen (for home photo voltaic system, in particular), and 3) automobiles. The methods and results of conjoint analysis for 1) home electric appliances at around 100,000 yen will be presented here.The analysis was conducted for 1,112 respondents who were sampled and selected statistically on the web. Attribute j was set as four attributes including initial price, annual electricity cost, product reliability, and environmental performance. Figure 6 is an example of the profile. The two profiles in the figure have different levels of “product price” and “environmental performance.” Twelve profiles were shown to each respondent who were asked to indicate his/her preference of each profile according to seven-step scale. Based on the responses, coefficient of preference weight (wij of equation (5)) for each attribute j of each respondent i was calculated. Each attribute was calculated as having the following values as average of respondents when converted to product price:・ Annual electricity cost is 1,000 yen less Same value as product price being 5,500 yen less・ Product reliability is high Same value as product price being 10,000 yen less02040608010019551960196519701975実績値Bassモデルによる近似曲線02040608010019551965196019751970Actual valueApproximated curve by Bass model① ② ③ ④ ⑤ ⑥ ⑦① ② ③ ④ ⑤ ⑥ ⑦Choice 2Environmental performanceAnnual electricity costProduct reliabilityProduct priceChoice 1Very attractiveDon’t knowNot attractive at allVery attractiveDon’t knowNot attractive at allHighAverage100,000 yen15,000 yenEnvironmental performanceAnnual electricity costProduct reliabilityProduct priceHighGood120,000 yen15,000 yen1964-2003118.10.120.133171958-200587.80.000420.182141959-200587.80.000530.182131970-2004100.50.00590.151131987-200676.50.0110.190131953-2005760.0210.099131961-200392.30.00690.148121961-200465.90.0410.145121992-2006112.20.0230.067121960-2003900.0110.120111964-2006118.10.00950.142101983-200477.80.000140.31391966-1981800.0380.26081987-200462.30.0120.53061955-1979101.20.0440.16561978-2004840.00420.41051955-198298.13.6x10-70.42951955-196895.90.0130.68141967-198299.20.000230.63831993-2007920.000420.7093NprCell phoneColor TVBlack & white TVRefrigeratorVTRWashing machineCD playerGas water heaterAutomatic washing machineFacsimileAutomobileWashing toilet seatStereoAir conditionerFluorescent lampPCMicrowave ovenAT car (light car)AT car (excluding light car)Flushing toiletSpeedPeriodProductFig. 5 Transition (actual) of household diffusion of washing machine and approximated curve by Bass model (p = 0.044, r = 0.165, N = 101.2).Fig. 6 Example of profile by conjoint analysis.Table 2 Diffusion coefficient values of past products.The source of diffusion transition data is taken mostly from Trend of Household Consumption – Annual Report of Consumption Trend Survey (for FY2004 and FY2006), Economic and Social Research Institute, Cabinet Office, Government of Japan. Rightmost column shows the years required for the diffusion to increase from 10 % to 50 %.

※このページを正しく表示するにはFlashPlayer9以上が必要です