Discipline Structure Optimization Of Higher Education Based On Educational System Engineering

Higher education is the highest level in the entire education system and itsdevelopment and reform are of great strategic significance to the education system. Inrecent years, China has made outstanding achievements in higher education, butmeanwhile there still exists some structure problem which has posed great challengeto the sustainable development of higher education. Thus, a dynamic adjustmentmechanism must be built to optimize higher education structure. Disciplinarystructure, whose optimization state directly determines the overall efficiency of highereducation structural optimization, is the heart of higher education structure. Based onthe current academic achievements, thought and methods in educational systemengineering are adopted in this dissertation to give dynamic assessment andoptimization to Chinese higher education disciplinary structure. The main innovativeexplorations of this doctoral dissertation are listed in the following four aspects.Firstly, multiple objective evaluation criteria of discipline structure in highereducation is established to give multiple objective dynamic evaluation to Chinesediscipline structure. According to the analyses of domestic and overseas disciplinestructure and optimization in higher education, a multi-dimensional objective ofdisciplinary structure assessment as well as a index system suitable to the multipleobjective evaluation in Chinese discipline structure are set up. Canonical correlationand cointegration model are utilized to make diachronical assessments of Chinesediscipline structure from2002.Secondly, a hybrid model which is available for linear and non-linearpredictability is constructed to predict the manpower demand for Chinese economicand social development. A hybrid model based on BPNN (Back Propagation NeuralNetworks) and ARMA (Auto-Regressive and Moving Average Model) is built to giveaccurate prediction to manpower demand. What’s more, taking the uncertainty offuture economic development into account, interval prediction is given to manpowerdemand according to the uncertainty forecast theory and non-parametric statisticalmethods to ensure the efficiency and reliability of mid-and-long term forecast formanpower demand.Thirdly, based on ARIMA (Autoregressive Integrated Moving Average) modeland BPNN, a hybrid forecasting model is made to predict the talent supply capability of Chinese discipline structure. With the comprehensive analyses of the factorsinfluencing the supply scale and full consideration of a restrictive condition, i.e.,reduction of school-age population for higher education, a mid-and-long term forecastis given to the total supply scale and separated supply scale of discipline structurewith the help of longer time series data.Lastly, Pareto optimality model is built to give multiple objective optimizations toChinese discipline structure and related policy suggestions are provided. Takenthoughts from educational system engineering, manpower demand and educationsupply forecasting into full consideration, disciplinary structural optimization can beelaborated in two parts, that is, general equilibrium solution and real feasible solution.In the analytical process of real feasible solution, based on the development state ofChinese discipline structure in various stages, the optimization of discipline structureis divided into two stages: the spontaneous regulation of market and governmentmacro adjustment and control as well as Pareto optimality, during which profound andcomprehensive analyses are provided respectively from the perspectives ofcomponent optimization and total optimization. Thus, respective policies andsuggestions of total optimization and component optimization are provided accordingto the different solution sets in the optimization process of Chinese disciplinestructure.