中国科学院数学与系统科学研究院期刊网

Systems Engineering - Theory & Practice 2022 Vol.42

Please wait a minute...
For Selected: Toggle Thumbnails
Population aging, immigration and economic growth
NI Xuanming, HE Yingjie, WU Kangping, PENG Fangping
Systems Engineering - Theory & Practice    2022, 42 (1): 1-12.   DOI: 10.12011/SETP2021-1185
Abstract1209)      PDF(pc) (1198KB)(725)       Save
The continuous imbalance of population structure brought about by population aging hinders economic growth. Immigration, to some extent, can directly supplement the labor force in the economic system. In order to explore the impact of immigration on aging economy, this paper constructs an economic model of aging based on immigration. Under certain conditions, this economic system has a unique saddle stable equilibrium growth path. On the saddle path, the role of immigration is "neutral", showing only a level effect rather than a rate effect on the economic system. Immigration improves the level of total capital, total output and social welfare on the equilibrium, but neither affects the growth rate of output or capital, nor alleviates the adverse impacts of population aging on economic growth. To make the immigration policy play a positive role in the background of aging, this paper further develops the rate of technological progress as a function of immigration, thus showing that introduction of immigrants with high human capital can offset the negative effects of population aging to a certain extent, so as to promote economic growth in the long run.
Reference | Related Articles | Metrics
The construction and application of seasonal weakening buffer operator
HE Lingyang, WANG Zhengxin
Systems Engineering - Theory & Practice    2022, 42 (1): 13-23.   DOI: 10.12011/SETP2020-2677
Abstract476)      PDF(pc) (784KB)(344)       Save
Given that the existing weakening buffer operators cannot effectively deal with the modeling and forecasting problems of seasonal shock disturbance systems, this paper constructs two types of seasonal weakening buffer operators, that is, seasonal average weakening buffer operator and seasonal full-information variable weight weakening buffer operator. On this basis, this paper further discusses the buffer strength and smoothness of the seasonal weakening buffer operator, and finds that they are better than that of the corresponding classical one. Finally, based on the grey wolf algorithm, this paper gives the weight optimization scheme of seasonal full-information variable weight weakening buffer operator and takes the impact and disturbance data prediction of the quarterly added value data of the secondary industry as an example to verify the effectiveness of the seasonal buffer operator. The forecasting results show that:For seasonal time series influenced by disturbance items, the adaptability and prediction accuracy of the seasonal buffer operator proposed in this paper is significantly better than that of the classic buffer operator. Further findings show that, compared with the SARIMA and EMD-ARIMA models, the prediction accuracy of two types of seasonal buffer operators is equivalent to that of the EMD-ARIMA model:Their average prediction relative error is about 3%, while that of the SARIMA model is as high as 15.65%.
Reference | Related Articles | Metrics
Research on financial risk spillover of the countries along the Belt and Road—Based on TENET method
ZHAO Wanli, FAN Ying, JI Qiang, ZHANG Dayong
Systems Engineering - Theory & Practice    2022, 42 (1): 24-36.   DOI: 10.12011/SETP2021-0432
Abstract1362)      PDF(pc) (1482KB)(808)       Save
With the continuous improvement of trade freedom and the cooperation level of financial institutions among the countries along the Belt and Road, the financial markets among these countries are also gradually integration. Research on identification, contagion and measurement of financial market risk among countries along the Belt and Road is of great practical significance to ensure the healthy development of regional finance. In this paper, the extreme risk spoillover network of stock markets along the Belt and Road is constructed by using TENET method, and the risk characteristics, risk sources, risk transmission paths and risk evolution laws of stock markets under extreme tail risk situation are explored. The research results show that the systemic risk index of the stock markets along the Belt and Road countries has time-varying characteristics, and presents an upward trend during periods of economic pressure. From a regional point of view, the European region was at high risk in 2008 due to the financial crisis, and the Asian region was at high risk during the 2020 COVID-19 pandemic. From the perspective of specific countries, Greece and Cyprus, which are more affected by the European debt crisis, are at higher risk. China mainly receives external financial risks in the Belt and Road financial risk network, which mainly come from Israel, Greece, Singapore and other countries. This research can provide theoretical guidance for macro policy makers and transnational financial investment institutions of countries along the Belt and Road to monitor financial risks and manage foreign imported risks.
Reference | Related Articles | Metrics
Research on credit risk contagion considering the interaction of relationships
QIAN Qian, ZHOU Yong, CHAO Xiangrui
Systems Engineering - Theory & Practice    2022, 42 (1): 37-45.   DOI: 10.12011/SETP2021-0033
Abstract557)      PDF(pc) (782KB)(586)       Save
In reality, there are two or more kinds of relationship between enterprises. Under the interaction of relationships, the path and intensity of credit risk contagion usually be changed, which makes the credit risk contagion more complex. Based on this, a dual-layer network is used to describe the topological structure of enterprise network within coexistence of asset and transaction relationship. A credit risk contagion model considering the interaction of relationships is constructed. And the influence of interaction on credit risk contagion is investigated. The results show that:Compared with the only asset relationship, the risk contagion threshold is smaller when the asset and transaction correlation coexist; the dual-layer network topology is an important factor affecting the contagion effect of credit risk; reducing the similarity of the dual-layer network can restrain the contagion of credit risk among enterprises.
Reference | Related Articles | Metrics
The study on inverse CIR bubble model
LIN Li, ZHENG Haitao, QIN Xiao
Systems Engineering - Theory & Practice    2022, 42 (1): 46-59.   DOI: 10.12011/SETP2021-0491
Abstract369)      PDF(pc) (1308KB)(357)       Save
According to the definition of stock price bubble, the detection of bubble ought to be a test of joint hypothesis, that is somehow confronted with logical predicament. Therefore, majority of scholars choose to by pass the definition and attempt to directly model the bubbles evolution as some proper stochastic processes, which are rationalized with the assumptions on the speculative behaviors. With the the statistical properties of models of the stochastic processes, some bubble detection methods are constructed accordingly. However, the existing stochastic process models have somehow strong restrictions on the nature of bubbles, which lead much cases which truly have bubbles are overlooked by the models. As a result, the sensitivity of the methods for detecting bubbles based on those models are lower. Besides, most models suffer from the lack of dynamics of bubble burst hazard rate, and thus lead to a unconvincing status for the real-time warning for upcoming bubble crash. To remedy traditional models' deficiency mentioned above, we propose a new bubble model in this paper. This model is called inverse CIR bubble model as the solution of stock price follows a reciprocal Cox-Ingersoll-Ross stochastic process. It has crystal clear economics implication as well as parsimonious mathematical specification with only three crux parameters. Hence the model is easier to make explanation and calibration. Based on the derived analytical results, this paper rationalizes the presence of nonlinear risk premium and transient super-exponential growth of price in a bubble. Meanwhile, this model can help us in explaining for some abnormal empirical results during bubble maturation, such as the "lull before the storm" and the "stagflation in high volatile plateau" before the bubble crash. These empirical results can however not reconcile with the traditional bubble models. Further, we show that this model has an endogenous hazard rate for the crash, which is of clear economics implication and can be used as a real-time warning indicator. Meanwhile, This paper prove the hazard rate satisfies a specific equation in the form of Gaussian hypergeometric function. As shown by the empirical studies on 2015 bubble for Chinese stock market, our model not only enjoys good explanatory power but also provides timely alarms for the upcoming collapse of the bubble.
Reference | Related Articles | Metrics
The theory and evidence of Chinese household risk asset allocation from the perspective of information uncertainty
WU Wensheng, LI Shuo, TAN Changchun, SHENG Shijie
Systems Engineering - Theory & Practice    2022, 42 (1): 60-75.   DOI: 10.12011/SETP2020-2867
Abstract405)      PDF(pc) (901KB)(641)       Save
This paper investigates the optimal allocation of household assets under rational inattention. We provide a model to explore how the information uncertainty affect asset allocation. Furthermore, we quantitatively analyze the relation between the reduction of information uncertainty and household portfolio by employing the data of CHFS. Meanwhile, we use the ability of information process and information acquisition as the reduction of information uncertainty. The study suggests that both of them can increase the possibility of households participating in the financial market and the proportion of households holding risky assets. The empirical results further find that the household with stronger information process abilities is more inclined to hold high-risk markets such as stocks. The relation between external information acquisition ability and the allocation of family risk assets is shown as an inverted U-shaped. The results also find that the own internal information processing ability and external information acquisition ability have a substitution effect in affecting the allocation of family risk assets.
Reference | Related Articles | Metrics
Will Sci-Tech innovation board improve the quality of China's stock market: Based on multi-agent modeling
LIANG Rui, DONG Jichang, HE Zhou, LIU Ying
Systems Engineering - Theory & Practice    2022, 42 (1): 76-83.   DOI: 10.12011/SETP2019-2688
Abstract1257)      PDF(pc) (839KB)(384)       Save
Sci-Tech innovation board has received extensive attention from market participants since its opening. Within the first 10 trading days, the stock price of Sci-Tech innovation board all doubled. The trading mechanism of Sci-Tech innovation board adopts the T+1 trading mechanism. There is no price limit in the first 5 trading days, and the price limit is 20% after the first 5 trading days. What effect did the introduction of Sci-Tech innovation board have on the quality of China's stock market? This paper adopts multi-agent modelling to build an artificial stock market model in line with China's real market, simulates the trading mechanism of Sci-Tech innovation board, and measures the quality of the stock market from three aspects:Price discovery efficiency, market liquidity and market volatility. The results show that compared with the main board, the trading system of Sci-Tech innovation board improves the price discovery efficiency of the market, reduces market liquidity, and exacerbates market fluctuations.
Reference | Related Articles | Metrics
Survey of consumer finance research
ZENG Yan, YANG Yating, XU Fengmin, ZHANG Chengyi
Systems Engineering - Theory & Practice    2022, 42 (1): 84-109.   DOI: 10.12011/SETP2021-1290
Abstract1358)      PDF(pc) (1855KB)(881)       Save
Based on explaining the research scope of consumer finance, this paper innovatively combines bibliometrics and text mining methods to systematically sort out the research context of consumer finance, rebuild the research system with five core topics, and provide possible directions for future research in this field. First, this paper uses CiteSpace to analyze the core journals in the Chinese and English databases from 1998 to 2020. Then, the article summarizes different research focuses of Chinese and English literature at different time periods, and indicates the current research hotspots. Second, the paper refines the key words of the industry research reports by Python text mining method, and proposes five new characteristics of consumer finance in the new development stage. Third, the article summarizes five core topics to reshape the research system of consumer finance, and explains the key research content of each topic in detail. Fourth, we put forward possible future research directions of consumer finance from five different aspects. This article hopes to broaden the research boundary of consumer finance, promote the development and progress of the industry, and provide some valuable suggestions for our country's demand-side reform and the establishment of a dual circulation development pattern.
Reference | Related Articles | Metrics
Optimal allocation policies for emergency medicine reserve methods against major public health events
ZHANG Weijian, SHI Xianliang, HUANG Anqiang, HUA Guowei, WANG Shouyang, LI Jiangning
Systems Engineering - Theory & Practice    2022, 42 (1): 110-122.   DOI: 10.12011/SETP2021-0620
Abstract788)      PDF(pc) (1286KB)(640)       Save
Global major public health events have become a serious challenge that endangers human life and destroys social and economic order. Under this situation, it can be seen that emergency medicine reserve is the basis and guarantee of implementing emergency rescue and life-saving. At the same time, emergency medicine reserve directly affects the speed and final effect of emergency response. However, there is no quantitative research on the emergency medicine reserve methods (physical reserve and capital reserve) in the existing studies. Thus, this paper determines the optimal allocation policies of physical reserve and capital reserve of emergency medicine under stochastic demand. Firstly, we establish the reserve allocation model by minimizing the long-term average cost, which quantifies four allocation policies. Besides, we establish the procurement and deployment model of emergency rescue stage, and discuss how to use medicine reserve. Finally, we analyze how the emergency medicine characteristics and demand characteristics affect the optimal physical and capital reserve, so as to adjust the reserve allocation policies more reasonably when the key factors change. The key insights include:1) Once a certain threshold for the penalty cost is passed, building up capital reserve becomes more cost-effective than building up physical reserve; 2) While the optimal capital reserve level always increases with the uncertainty of the demand, the optimal physical reserve level may decrease with it. The research findings can provide decision support for relevant government departments to respond to public health events.
Reference | Related Articles | Metrics
The optimal joint incentive contract designs and inventory decisions with an overconfident agent
KONG Xiangyin, LIU Shuqi, SHEN Xiaobei, FENG Gengzhong
Systems Engineering - Theory & Practice    2022, 42 (1): 123-137.   DOI: 10.12011/SETP2021-0660
Abstract523)      PDF(pc) (848KB)(427)       Save
This paper investigates the principal's joint incentive contract designs and inventory decisions when the agent is subject to the behavioral bias such as overconfidence. Building on the principal-agent model and newsvendor model, we distinguish two common types of overconfidence:Overestimation and overprecision, and investigate their impacts on the joint optimal contract designs and optimal inventory decisions, respectively. We find that quota-bonus contracts that clear the inventory is optimal and different types of overconfidence can lead to significantly different or even opposite insights on the optimal decisions. Specifically, overprecision (resp. overestimation) results in a higher (resp. lower) bonus of the optimal contract and a lower (resp. higher or lower) inventory level compared with the counterpart with a rational agent. Furthermore, overestimation may yield higher profit than the first-best solution without moral hazard and mitigate the well-known overstocking phenomenon caused by moral hazard while overprecision may exaggerate the overstocking behavior.
Reference | Related Articles | Metrics
A structural lower bound of online task allocation for sharing platforms
DAI Wenqiang, JIANG Yuqi
Systems Engineering - Theory & Practice    2022, 42 (1): 138-143.   DOI: 10.12011/SETP2020-1843
Abstract287)      PDF(pc) (491KB)(253)       Save
In the process of task allocation on sharing platforms, the following situations are often encountered:The decision-makers need to decide how to reasonably allocate the current demand to the existing server to maximize the profit of the platform, when the future demand task sequence information (arrival time, start time, duration, etc.) is unknown. There is a limit on the number of servers on the platform, and at the same time, it cannot be re-allocated once a demand is allocated. The models and algorithms established previously mainly used for static task allocation environment, but here we need a dynamic task allocation model with above constraints. This paper establishes an online sharing platform task allocation model with a maximizing platform's profit objective, where the profit include not only variable proportion for the shares but also the fixed income. Applying the Yao principle, we give a lower bound of competitive ratio for this problem. This lower bound does not need any complexity assumption, so it is a structural lower bound.
Reference | Related Articles | Metrics
The impact of bounded rational differences in consumers onsuppliers' product pricing: A two-level population game model
WANG Zhaohua, LI Hao, ZHANG Bin
Systems Engineering - Theory & Practice    2022, 42 (1): 144-154.   DOI: 10.12011/SETP2020-2900
Abstract533)      PDF(pc) (976KB)(500)       Save
Affected by the difference of bounded rationality, consumers will adopt different methods such as pure imitation or probabilistic imitation when learning advantage strategies, and have an important impact on the supplier's product pricing strategy. Based on population game modeling and simulation technology, this paper analyzes suppliers' optimal product prices under two forms of bounded rationality, pure imitation and probabilistic imitation, and further discusses the impact of product costs and transaction costs on product prices. The research results show that:1) compared with pure imitation, the probabilistic imitation of consumers will prompt suppliers to give higher product prices; 2) the differentiation of supplier product costs and the decline of market transaction costs will depress product prices. The research conclusions are of great significance for guiding consumers to make reasonable decisions and improving consumer welfare.
Reference | Related Articles | Metrics
Research on the influence of variable investment amount mechanism on the voluntary public goods game
WANG Shuai, XU Zhaojin, ZHANG Lianzhong
Systems Engineering - Theory & Practice    2022, 42 (1): 155-171.   DOI: 10.12011/SETP2020-1632
Abstract477)      PDF(pc) (2396KB)(260)       Save
Public goods play an important role in social development. Private voluntary supply of public goods has always been one of the hot topics in academia. Based on the consideration that the amount of public goods investment in reality has individual heterogeneity and can be adjusted dynamically, this paper introduces the variable investment amount mechanism in the traditional fixed investment voluntary public goods game. Individuals can not only choose whether to provide public goods, but also decide how many public goods to provide. This more realistic model makes the game results have more explanatory power to the real problems. The results of this paper show that there is a double-edged sword effect on the improvement of social cooperation level by the mechanism of variable investment amount, which is mainly reflected in that only when the rate of return is moderate, the proportion of cooperators in the game equilibrium of variable investment will be higher than that of fixed investment. However, the variable investment amount mechanism can still improve the average social income and the total supply of public goods in most cases, which alleviates the tragedy of the commons to a certain extent. It is found that increasing the upper limit of investment amount will lead to a small increase in the proportion of cooperators and then a decrease in it. The increase of the income of the loners has an inverted U-shaped effect on the level of social cooperation, which is mainly related to the change of the income difference of the three strategies. Besides, in the process of adjusting the mixed strategy, the appropriate response intensity to the last round of income is conducive to the emergence of social cooperation and the improvement of cooperation level. This study provides a new perspective and theoretical basis for the private voluntary supply of public goods in reality.
Reference | Related Articles | Metrics
Is the subcontractor's altruism beneficial?—Take commissioned research and development of public good technological innovation as an example
WU Baiyu, ZHONG Weijun, MEI Shu'e
Systems Engineering - Theory & Practice    2022, 42 (1): 172-183.   DOI: 10.12011/SETP2020-1640
Abstract649)      PDF(pc) (1017KB)(395)       Save
To clarify whether the subcontractors' altruistic preferences are conducive to the public sector, this paper constructed a principal-agent model to study the influence of subcontractors' altruism on the commissioned research and development strategy of public goods technological innovation under the stochastic dependencies. The results show:Stochastic dependence and altruism have the same force, which improves the advantage of relative performance evaluation. The altruism of subcontractors is not necessarily conducive to the increase in the expected benefits of the public sector. The public sector does not want both of his subcontractors to behave equal intensity of altruistic preference. Instead, public sector can only profit when the subcontractors are asymmetrical, and the stronger the intensity of altruism of contributor with low capability, the more beneficial to expected benefits. The public sector can benefit from the difference in the strength of the altruism, but pay attention to the matching of preference and capability.
Reference | Related Articles | Metrics
Renewable clean energy and clean utilization of traditional energy: An evolutionary game model of energy structure transformation of power enterprises
CHAI Ruirui, LI Gang
Systems Engineering - Theory & Practice    2022, 42 (1): 184-197.   DOI: 10.12011/SETP2021-0149
Abstract744)      PDF(pc) (977KB)(645)       Save
The clean use of traditional energy and the incorporation of renewable clean energy have become the focus of the low-carbon energy structure transformation of power enterprises. Considering the "two-wheel drive" of renewable energy and the clean utilization of traditional coal, we construct an evolutionary game between government departments and power enterprises, analyzes the evolutionary stable equilibrium of government subsidy policies and power enterprises' choice of energy structure, and discusses the boundary conditions for the inclusive development of the clean use of traditional energy and renewable clean energy. The results show that there are transitional states such as periodic solutions or the government taking non-subsidy and the enterprise taking integrated energy power generation states in the evolution process of transitioning to the clean power generation mode. With the support of government preferential policies, enterprise reduces the cost of clean technology innovation and promotes low-carbon energy structure transformation. In addition, the ratio of renewable clean energy to traditional power capacity depends on the expected benefits of different energy generation, types of government subsidies and their environmental benefits.
Reference | Related Articles | Metrics
Shift scheduling and rostering problem of call centers with part-time shift consideration
WANG Xiuli, WANG Ruicheng, HU Xiuwu
Systems Engineering - Theory & Practice    2022, 42 (1): 198-210.   DOI: 10.12011/SETP2020-0065
Abstract359)      PDF(pc) (888KB)(316)       Save
In recent years, call centers incorporate part-time manpower into their schedule tables. Part-time manpower is different from full-time agents in cost and work mode. An integer programming model is constructed under this actual scenario. A two-stage optimal algorithm is proposed based on analyzing and recognizing the optimization structure of the problem. The algorithm adopts the idea of decomposed decisions of day and shift on-duty to reduce the complexity for solving the problem, and assigns full-time and part-time shifts by using the different metrics that cover human resource requirements of intervals to generate available solution. Then, the algorithm uses neighborhood search techniques with simulated annealing mechanism or greedy mechanism to improve the generated solution. The computational experiments illustrate that the proposed algorithm can obtain optimization schemes of the enterprise-scale instances. Finally, how various factors affect optimization strategies of part-time manpower is analyzed. The research shows that scientific and rational use of part-time manpower can effectively reduce the cost of schedule tables.
Reference | Related Articles | Metrics
Causal inference for unbalanced cases in incomplete data with doubly robust estimators
YU Haiyan, XIANG Jiao, GAO Mingyue
Systems Engineering - Theory & Practice    2022, 42 (1): 211-223.   DOI: 10.12011/SETP2020-0268
Abstract566)      PDF(pc) (785KB)(464)       Save
Heterogeneity and missingness are two essential issues in causal inference with massive data. This paper presents a model for estimating the causal effect with double robustness through re-randomizing the treatment assignment and obtaining de-biased estimates of parameters. We first clarified the causal inference with the identifiable and estimability in observational data. We constructed an integer optimization model for minimizing absolute standard bias with the distribution of assignment probability. This model reduces the covariance imbalance between the two groups. The selected data subsets were obtained from the optimal solution or approximate optimal solution of the model, which randomized the two groups' cases. An augmented inverse probability weighted estimator was implemented to estimate the missing response of observations. The results were doubly robust for both the uncertainty from the estimation of presences and the assignment probabilities. Finally, we validated the efficacy of this method through real cases and extended numerical analysis.
Reference | Related Articles | Metrics
Decision-making model and dynamic simulation analysis of the development and utilization of “urban mining” from a multi-agent perspective
CHEN Liqiang, GAO Ming, LIANG Kairong
Systems Engineering - Theory & Practice    2022, 42 (1): 224-240.   DOI: 10.12011/SETP2020-2570
Abstract480)      PDF(pc) (1428KB)(618)       Save
To investigate the effects of different subjects' behavioral decisions on the development and utilization of "urban mining", this paper starts from five main subjects:Residents, public institutions, classification units, resource recycling enterprises, and the government. Firstly, the system dynamics model of the development and utilization of "urban mining" is constructed using the grey model theory approach. The validity of the model is judged by the method of qualified verification of the degree of incidence. Secondly, we conduct mid- and long-term dynamic simulation of the five main subjects' behavior decisions. The results show that the trading platform and value recognition have the most significant impact on residents and public institutions' behavior. Waste circulation information has the most significant effect on the behavior of sorting units. For resource recycling enterprises, the supply and demand of renewable products is the main factor in their behavior changes. Furthermore, by further analyzing the external environmental factors of "urban mining" recovery, it can be seen that a single policy has a limited standardized effect on "urban mining" recovery. On the contrary, optimizing formal recycling channels, regulating informal recycling channels, and inhibiting illegal dumping are conducive to developing and utilizing "urban mining". Finally, by comparing the five significant subjects' combined plans, it indicates that improving residents' and public institutions' behavior decisions is key to developing and utilizing "urban mining".
Reference | Related Articles | Metrics
Dynamic synchronization analysis method of operation system-of-systems based on target stream drive
LI Sen, WANG Liang, CHEN Gang, WU Jing
Systems Engineering - Theory & Practice    2022, 42 (1): 241-252.   DOI: 10.12011/SETP2019-2739
Abstract400)      PDF(pc) (1515KB)(538)       Save
Based on the facts, the time-varying characteristics of node-coupling relevancy, can't be reflected effectively by the static-synchronization analysis for operation system-of-systems (SoS) under fixed structure, another dynamic-synchronization analysis method for operation SoS, driven by target stream, is proposed. Regarding target's attacking as the essential impetus to the SoS response and structural adjustment, the positive and negative effects of target stream on synchronization are discussed, by analyzing the macro characteristics on element attributes and topology evolution under the condition of SoS combat. Integrated with combat ring theory and coupled oscillator model, the mechanism and characterization of SoS phase synchronization are studied. Moreover, the updating method of node phase iteration is presented. Compared with the static-synchronization, the simulation results exhibit obvious differences among the dynamic-synchronization possessing special dynamic characteristic, and the oscillation pattern and synchronization ability.
Reference | Related Articles | Metrics
Forward variable selection method based on k-nearest neighbor mutual information and its application in soft sensor modeling of water quality parameters
WANG Wei, YANG Chunhua, HAN Jie, LI Wenting, LI Yonggang
Systems Engineering - Theory & Practice    2022, 42 (1): 253-261.   DOI: 10.12011/SETP2020-2882
Abstract1508)      PDF(pc) (803KB)(330)       Save
Soft sensor technology realizes real-time prediction of difficult-to-measure variables by constructing a mathematical model between easy-to-measure auxiliary variables and difficult-to-measure primary variables. In order to effectively analyze the correlation and redundancy between variables and realize the selection of auxiliary variables, this paper proposes a forward variable selection method based on k-nearest neighbor mutual information. Based on the criterion of maximizing the forward cumulative mutual information value of input variables to select correlated variable, and the redundant mutual information value between each new added variable and the subset of selected variables is calculated to judge whether the added variable is redundant variables. By setting the threshold of redundant mutual information value, the redundant variables are eliminated, and the optimal subset of auxiliary input variables can be obtained. The simulation results based on a numerical case verify the feasibility and effectiveness of the variable selection method proposed in this paper, which not only accurately selects auxiliary variables but also reduces the complexity of the algorithm. Finally, the method was successfully applied to the selection of input variables for the effluent biochemical oxygen demand (BOD) prediction model in the wastewater treatment process, and the selected auxiliary variables were used to effectively improve the prediction accuracy of the model.
Reference | Related Articles | Metrics
Research on modeling and analysis of BeiDou satellite navigation system effectiveness
XUE Yu, YANG Zhuopeng, YANG Yixin
Systems Engineering - Theory & Practice    2022, 42 (1): 262-272.   DOI: 10.12011/SETP2020-2928
Abstract489)      PDF(pc) (1213KB)(665)       Save
For the multi-factors and complex relationship of navigation system effectiveness, an index system is established based on quality characteristic, and an integrated method of system effectiveness analysis is proposed, considering the mean time between outages, satellite in-orbit testing time, launch vehicle reliability, relationship of satellites as well as signal-in-space accuracy. The single navigation satellite availability is confirmed based on generalized stochastic Petri Nets, the multi parameters reliability is analyzed considering the random faults and wearout faults, the performances are modelled based on stress-strength analysis, and the navigation system effectiveness is presented based on dynamic Bayesian networks, considering relationship of each satellite. The method is verified by simulation of BeiDou satellite navigation system, and can be instructive for the system effectiveness study of others satellites.
Reference | Related Articles | Metrics
Systems Engineering - Theory & Practice    2022, 42 (10): 0-0.  
Abstract295)      PDF(pc) (16881KB)(94)       Save
Related Articles | Metrics
The labor leverage discount effect in China: An explanation from labor-technology substitution
YIN Libo, WEI Dong
Systems Engineering - Theory & Practice    2022, 42 (10): 2565-2588.   DOI: 10.12011/SETP2022-0200
Abstract564)      PDF(pc) (1135KB)(308)       Save
Under the dual background of transformation and upgrading of Chinese economy and the new round of technological revolution, we select the quarterly data of Shanghai and Shenzhen A shares from 2011 to 2019 to investigate whether there is a substitution effect between labor factor and technology factor based on the micro perspective, and then provide an explanation for the labor leverage discount in China. We find that: 1) There is a significant substitution effect between the firm's labor factor and technology factor, and the degree of substitution is more obvious in the high-tech industry, state-owned firms and firms with low financing constraints; 2) The performance of this effect is different in different economic conditions: The substitution effect is stronger during the economic downturn, which indicates that the best time for firm to make technology substitution decision is economic recession; 3) Labor-technology substitution can significantly weaken the negative effect of labor leverage on stock returns. This result is still robust after replacing the labor technology-substitution variable with total factor productivity and labor productivity, which can also directly reflect the the firm's technical level. The result prove that the firm's technical level has a significant inhibitory effect on labor leverage discount, indicating that labor factor can effectively promote the positive effect of low-level labor leverage on stock returns. Related results can provide ideas for firms to take effective measures to cope with the rising labor costs and the shortage of labor supply caused by the aging population. At the same time, those results also provide empirical support for the in-depth understanding of the pricing mechanism of labor leverage in Chinese A-share market, the rational allocation of production factor, the transformation of the firm's growth mode and the optimization and upgrading of the firm's production structure.
Reference | Related Articles | Metrics
Shadow banking, liquidity stratification, and policy interest rate transmission
NI Xuanming, WANG Jiangwei, ZHAO Huimin
Systems Engineering - Theory & Practice    2022, 42 (10): 2589-2602.   DOI: 10.12011/SETP2022-1188
Abstract515)      PDF(pc) (1033KB)(239)       Save
The monetary policy framework of the People's Bank of China (PBOC) points out that it is necessary to guide the trend of market interest rates through the policy interest rates system. However, the negotiable certificate of deposit (NCD) interest rate often deviates greatly from the medium-term lending facility (MLF) interest rate. By constructing the policy interest rates transmission model under the control of credit interest rates, this paper puts forward three influential factors of spreads between market interest rates and policy interest rates: Shadow banking, inter-bank lending market financing difficulty, and legal deposit reserve ratio. Liquidity stratification is the external embodiment of the latter two factors, while shadow banking can be divided into traditional shadow banking and bank shadow with opposite roles. In terms of economic mechanism, traditional shadow banking and loans are independent and complementary, while bank shadow is an alternative loans for regulatory arbitrage of China's commercial banks. In addition, the overall liquidity abundance of the interbank market and the monetary policy transmission efficiency caused by the liquidity gap between financial institutions can explain the impact of liquidity stratification.
Reference | Related Articles | Metrics
Research on the system risk spillover effects among crude oil, gold, estate and financial sectors in China
DAI Zhifeng, ZHU Haoyang, YIN Hua
Systems Engineering - Theory & Practice    2022, 42 (10): 2603-2616.   DOI: 10.12011/SETP2021-2322
Abstract485)      PDF(pc) (1044KB)(355)       Save
As an open system, the stability of the financial system is affected by many markets. It is of great significance to study the infection mechanism between financial markets, or between financial departments and the other markets to maintain the stability of the financial system. This paper uses value at risk (VaR) to measure the tail risk, and investigates the risk contagion effects among China's crude oil, gold, real estate and four financial sectors, adopting the variance decomposition spillover index framework based on TVP-VAR model. The results indicate that: 1) the total spillover index (TSI) among the analyzed assets is as high as 81.37%, suggesting that tail loss is highly infectious in the financial system. 2) crude oil and bank are the largest net transmitter and the largest net receiver of the system shocks, respectively. The real estate industry has the largest positive net spillover effect on the banking sector. 3) Market volatility (VIX) and term spread (TS) have strong explanatory ability to the total spillover index (TSI). Investors and regulators should pay full attention to the important role of stock and bond markets in systemic risk early warning.
Reference | Related Articles | Metrics
Imbalanced corporate bond default modeling using generative adversarial networks oversampling techniques
YAO Xiao, LI Ke, YU Le'an
Systems Engineering - Theory & Practice    2022, 42 (10): 2617-2634.   DOI: 10.12011/SETP2021-2328
Abstract400)      PDF(pc) (1089KB)(216)       Save
Based on the data of corporate bond issuers in Chinese market, this study applies oversampling techniques including Wasserstein generative adversarial networks (WGAN) and SMOTE to the imbalanced sample to improve the performance of bond default prediction. To explore the effect of oversampling techniques on classification models, the predictive outputs with difference imbalanced ratios are reported in the experimental results. It finds that the classification performance is significantly improved with the application of oversampling techniques, and the improvement is further enhanced when the sample distribution becomes more balanced. Compared to the classical SMOTE technique, both AUC and F1 score can be improved by WGAN. Overall, the experimental results demonstrate that the predictive performance of bond default models can be effectively boosted by generating artificial minority samples based on WGAN combined with the application of machine learning algorithms, which provides new insights into the bond default risk prediction of imbalanced samples.
Reference | Related Articles | Metrics
Optimal option portfolio with time-varying volatility
ZHOU Chunyang, WU Chongfeng
Systems Engineering - Theory & Practice    2022, 42 (10): 2635-2643.   DOI: 10.12011/SETP2021-2201
Abstract261)      PDF(pc) (651KB)(215)       Save
In this paper, we develop a one-period model to solve the optimal option portfolio problem for SSE 50ETF options. We use the GARCH model and GJR-ST model to describe the dynamic process of the daily logarithmic return of the underlying 50ETF, and use Monte Carlo simulation method to generate the price distribution of the underlying on the option maturity date. The out-of-sample empirical results show that the option portfolio constructed by GARCH model and GJR-ST model can both obtain significantly positive out-of-sample average returns. Meanwhile, compared with GARCH model, GJR-ST model can better manage the volatility risk, so it can obtain higher return-risk ratios.
Reference | Related Articles | Metrics
Aggregating dynamic expert advice for online portfolio selection based on myopic correction
ZHANG Yong, HUANG Menghu, LONG Wanrong, YANG Xiaoguang
Systems Engineering - Theory & Practice    2022, 42 (10): 2644-2656.   DOI: 10.12011/SETP2020-2448
Abstract259)      PDF(pc) (949KB)(131)       Save
During the long-term investment process, the data with a longer time in the past have less influence on the current decision-making, and the investor often makes decisions by referring to the recent information. Considering this actual situation, based on the latest stock price information, this paper uses an exponential smoothing method to gradually correct the myopia of static expert advice, and proposes an online portfolio strategy integrating dynamic expert advice. In order to better fit the actual situation, this paper further considers transaction costs to obtain strategy based on integrating dynamic expert advice. It is proved theoretically that the cumulative gains of online portfolio strategies perform as well as the best expert advice, having a good competitive performance. Using real stock market data for numerical example analysis, the result shows that when the transaction cost is zero or non-zero, the proposed online portfolio strategies can track the best expert advice in terms of cumulative wealth. Compared with the related strategies, the proposed strategies achieve better gain. Meanwhile, they own similar performance in terms of risk and risk-adjusted return, and they are also sensitive to transaction costs in a similar extent.
Reference | Related Articles | Metrics
Estimating risk measures via stochastic Kriging with stylized model and gradient information
CAI Xiaoting, YUN Xin, JIANG Guangxin
Systems Engineering - Theory & Practice    2022, 42 (10): 2657-2676.   DOI: 10.12011/SETP2021-3268
Abstract326)      PDF(pc) (1342KB)(120)       Save
Nested simulation is commonly used in estimating the risk measures of a portfolio with financial derivatives. But nested simulation requires a large number of simulation budgets and has low computational efficiency. This paper proposes a new method, i.e., the stochastic Kriging (SK) with stylized model, to fit the loss function of the portfolio instead of a large number of inner simulation budgets in the nested simulation, thereby improving computational efficiency. Moreover, we propose two ways to embed gradient information into SK with stylized model, i.e., embedding the gradient information directly into SK and using gradient information for interpolation. The latter is more flexible in choosing the stylized model. By the asymptotic normality of the maximum likelihood estimation, this paper derives a statistical hypothesis test to evaluate the effectiveness of the stylized model with gradient information. In the numerical experiments, we use examples of European options and Asian options based on geometric Brownian motion, European options based on the normal inverse Gaussian process, and portfolio with multiple options to verify the effectiveness of the proposed methods. The results show that the SK method with stylized models and gradient information can improve estimation accuracy and increase computation efficiency.
Reference | Related Articles | Metrics
Hedging strategy of long-term oil and natural gas contracts: A theoretical and empirical perspective from discretization
ZHAO Xinwei, MA Chaoqun, YUE Shengjie
Systems Engineering - Theory & Practice    2022, 42 (10): 2677-2696.   DOI: 10.12011/SETP2021-1845
Abstract252)      PDF(pc) (970KB)(106)       Save
Oil and natural gas are playing a fundamental role in global economy, as the most important country of energy consumption, China imports oil and natural gas mainly depend on long-term contracts. Traditional strategies are not suitable for hedging these risks, because they can not control the uncertainty of cashflow while hedging. Based on the assumption of the underlying asset follows an O-U stochastic process, a discrete model is established for measuring multi-dimensional risks with short-term futures contracts, high-dimensional Newton's method is designed for solving the discrete optimization problem, by which the optimal hedging strategy can be approximated efficiently, and the convergence of the algorithm is proved in this paper. Numerical result reveals that the optimal hedging strategy can be characterized by “three steps”. Empirical results on crude oil, heating oil and natural gas markets reveal that, the optimal hedging strategy under O-U process dominate others on cash-flow risk, terminal spot risk ratio and the cost of hedging, these demonstrate that our method perform well and is easy for application in practice.
Reference | Related Articles | Metrics
Research on the exposure-lag-response relationship between the new LNG pricing pattern and the supply-demand imbalance in the natural gas market
CHAI Jian, ZHANG Xiaokong, ZHANG Xuejun, WANG Yabo
Systems Engineering - Theory & Practice    2022, 42 (10): 2697-2709.   DOI: 10.12011/SETP2021-2601
Abstract319)      PDF(pc) (1575KB)(120)       Save
In response to the phenomenon of inadequate matching between traditional LNG pricing mechanism and the supply and demand laws of natural gas market, a comparative analysis of the effects of traditional and new LNG pricing patterns on the imbalance between supply and demand in the natural gas market is of great significance for price mechanism improvement. This paper portrayed the degree of matching between variables in three dimensions: Time series similarity, amplitude similarity and period similarity, and used the DLNM model to describe the exposure-lag-response relationship between variables. The results showed that the amplitude fluctuations and cyclical evolution patterns of coal price and supply-demand patterns in the natural gas markets were most similar; coal-gas linkage and oil-gas linkage were more suitable for China's natural gas market supply-demand patterns than gas-gas linkage. In terms of the one-day lagged effect, low oil and coal prices had a “protective effect” on the imbalance between supply and demand in the natural gas market; the injurious effect of high oil prices was not sustainable, while the opposite was true for high coal prices. In terms of cumulative lagged effects, the cumulative effects of oil, coal, Henry Hub and Asian LNG prices on the imbalance between supply and demand in the natural gas market showed inverted W-shaped, tick-shaped, ramp-shaped and bell-shaped patterns respectively; the injurious effect of oil price on the imbalance between supply and demand in the natural gas market peaked at 89.5/barrel, while the effect of coal price was least at 65/ton. The results provided important reference for future natural gas price reform.
Reference | Related Articles | Metrics
International crude oil price forecasting with news influence index attenuation
LIU Ling, WANG Jujie, LI Jianping
Systems Engineering - Theory & Practice    2022, 42 (10): 2710-2720.   DOI: 10.12011/SETP2021-2010
Abstract268)      PDF(pc) (1258KB)(157)       Save
With the development of big data technology, news data has been applied to forecast international crude oil price, but there is a lack of research on the impact degree and duration of news data. In order to quantify the decay of news influence, this paper puts forward a new exponential decay analysis method based on exponential decay and mutual information. In order to improve the prediction accuracy, this paper has constructed a combined kernel function support vector regression prediction model based on differential evolution optimization algorithm which realizes the optimal selection of weight coefficient, kernel function parameters and regression model parameters. To evaluate the effectiveness of the proposed methods, eight models are selected for comparative study. The empirical results show that integration of news influence decay time series can improve the correlation and prediction accuracy of crude oil price; the proposed forecasting method has good prediction accuracy, the average absolute percentage error of verification set is 1.53%, which is better than the comparison models.
Reference | Related Articles | Metrics
Research on collaborative distribution problems and cost allocation methods considering differences in service quality of logistics enterprises
RAO Weizhen, MIAO Xiaohe, ZHU Qinghua, JIANG Liwen
Systems Engineering - Theory & Practice    2022, 42 (10): 2721-2739.   DOI: 10.12011/SETP2021-3282
Abstract367)      PDF(pc) (1238KB)(178)       Save
The traditional collaborative distribution cost allocation assumes that there is no difference in the quality of services across all alliance enterprises, and takes the marginal contribution as the basis. However, there are obvious differences in the service quality of enterprises during the practical cooperation process. On this basis, this study first proposes a hierarchical framework for service quality evaluation under a collaborative circumstance, which includes four dimensions: Relationship performance, operation performance, cost performance and collaborative performance. Then, a multi-owner collaborative delivery vehicle routing problem model is constructed given the customer time window. Combined with the solution rules of the classical Shapley value method, the process of adjusting the cost allocation results is proposed as well. Finally, this paper verifies the effectiveness of the adjustment process by numerical experiments. The results show that: 1) According to the actual data, if SF Express, ZTO Express, YTO Express, and STO Express cooperate, the neglect of the difference in service quality among enterprises will cause a cost deviation of about 16.38%, leading to unfair cost allocation; 2) Vehicle input amount, on-time delivery rate and distribution price are particularly important in the evaluation of collaboration service quality; 3) Alliance members would like to narrow the gap in service quality to avoid the punishment losses, finally promoting overall service quality of the alliance. The adjustment process proposed in this paper can objectively reflect the influence of differences in service quality among enterprises on the cost allocation results. Enterprises with higher service quality can save more costs, while enterprises with lower service quality must bear the corresponding punishment to help obtain the fair allocation, thus achieving the purpose of encouraging members to improve service quality.
Reference | Related Articles | Metrics
Research on seeding marketing strategy considering the observation and experience characteristics of new product
MAO Zhaofang, ZHANG Yueqing
Systems Engineering - Theory & Practice    2022, 42 (10): 2740-2756.   DOI: 10.12011/SETP2021-1540
Abstract365)      PDF(pc) (1464KB)(150)       Save
Under the background of online and offline phased sales, this paper studies the seeding strategy of a monopoly seller who eliminates consumers' uncertainty about new product and influences other consumers through ratings by distributing free trial. This paper establishes a seeding model considering the product's observation and experience characteristics, analyzes the optimal seeding amount and income of seller under different pricing strategies and market environments, and further studies the model of seller controlling seeding influence. The results show that: 1) whether the seller seeds or not is related to the positioning of offline target consumers, the distribution of consumers, and the proportion of experience characteristics; 2) the benefit of seeding is more obvious when serving all potential consumers offline, and the larger the proportion of consumers with high value or the more products tend to experience products, the larger the seeding scope should be; 3) the seller shouldn't seed when serving only high valued consumers offline better; 4) when seller decide the seeding influence and serve all consumers offline, the more product tends to experience good, the more motivated the seller is to improve influence and reduce seeding to increase the consumers affected by seeding.
Reference | Related Articles | Metrics
Online product reviews helpfulness prediction based on topic analysis
ZHANG Wen, WANG Qiang, DU Yuhang, NIE Kun, LI Jian
Systems Engineering - Theory & Practice    2022, 42 (10): 2757-2768.   DOI: 10.12011/SETP2021-1206
Abstract604)      PDF(pc) (823KB)(304)       Save
With the prosperity of e-commerce, online reviews have become an important information source for both the online consumers and vendors in their decision making. However, with the marvelous reviews on the e-commerce platform, it is hard for consumers and vendors to acquire valuable information to support their decision making. This paper proposes a novel topic model called help-LDA by extending the classic LDA model with considering the helpfulness of online reviews. On the one hand, the proposed help-LDA model can extract helpful topics from online reviews. On the other hand, the proposed help-LDA model can be used for online review representation with goal of predicting the helpfulness of online reviews. With the real data collected from Dianping.com, we conduct extensive experiments to compare the proposed help-LDA model and the baseline models in topic modeling and helpfulness prediction of online reviews. The experimental results demonstrate the superiority of the proposed help-LDA model over the baseline models.
Reference | Related Articles | Metrics
The impact of feedback on users' follow-on innovation contribution in firm-sponsored online innovation community—The moderating effect of new product development phase
LIU Wei, ZHANG Boliang
Systems Engineering - Theory & Practice    2022, 42 (10): 2769-2781.   DOI: 10.12011/SETP2021-2220
Abstract449)      PDF(pc) (900KB)(152)       Save
The firm-sponsored online innovation community is an online platform through which firms cooperate with users to develop new products by value co-creation. It is a typical IT-driven business open innovation model. We propose a theoretical framework that the impact of feedback on users' follow-on innovation contribution in the firm-sponsored online innovation community based on social cognition theory, Yale attitude change theory and situational regulatory focus theory. Moreover, we build panel data model and test our theoretical hypotheses empirically using the grounded dataset in the MIUI community of Xiaomi. Our findings indicate that there is a significant positive relationship between positive feedback and users' innovation contribution, while there is also a significant correlation between negative feedback and users' innovation contribution. Considering different feedback sources, there are significant differences in the impact of feedback on users' innovation contribution. The results show that compared with peer feedback, expert feedback has a stronger impact on users' continuous innovation contribution. Further, we investigate the moderating effect of new product development phase on the relationship between feedback and users' innovation contribution. Compared with the idea generation stage, the negative feedback of experts in the testing and release stage has a positive impact on users' innovation contribution. Our findings contribute user innovation, new product development and open innovation literature in the context of online innovation community, and have some potential implications for product innovation practice in the emerging platform economy era.
Reference | Related Articles | Metrics
The staggering policy research in morning commute considering the travel mode of household travelers
HE Dongdong, ZHONG Qiuyan, QU Yi, LIN Zhengkui
Systems Engineering - Theory & Practice    2022, 42 (10): 2782-2796.   DOI: 10.12011/SETP2021-0402
Abstract291)      PDF(pc) (1290KB)(140)       Save
During the morning commute, the uncertainty of the travel mode choice of households increases the difficulty of formulating the staggering policy. Taking households as the research object, the travel cost model is constructed with the aim of minimizing the household travel cost by introducing the staggering policy and ridesharing coefficient. Based on the single-bottleneck network, a bi-level model considering the household travel mode and staggering policy is proposed. The results show that when the ridesharing coefficient is small, the staggering policy is not effective to reduce the travel cost of the household and the total system cost by changing the travel mode. When the ridesharing coefficient is large, the reduction rate of total system cost rises with the increasing ridesharing coefficient. Properly design of the staggering policy can effectively change the household travel mode and significantly reduce the travel cost of the household. Finally, the thresholds of the optimal staggering strategies under different scenarios are given, and the validity of the conclusions is verified by numerical examples, which can assist the decision-making of traffic congestion alleviation in morning commute.
Reference | Related Articles | Metrics
Healthcare referral strategies and coordination contracts based on hospital association service mode
LI Zhongping
Systems Engineering - Theory & Practice    2022, 42 (10): 2797-2814.   DOI: 10.12011/SETP2021-1062
Abstract370)      PDF(pc) (1448KB)(170)       Save
Based on the situation that promoting the effective development of hierarchical healthcare services takes hospital association as the carrier, this paper studies contractual schemes of cooperation between the government and medical institutions to coordinate the implementation of the referral service as well as the influence mechanism of the operations of hospital association service. By developing a four-stage game-theoretic queueing model, this paper analyzes the dynamic interaction relationships among patients, multiple community hospitals and tertiary hospitals in hospital association, and government, and then present all participants' equilibrium decisions in terms of patient referral rate, multiple community hospitals' capacity planning, and government subsidy price and each community hospital's profit-sharing price to a tertiary hospital. Our findings of studies indicate that whether the government coordinates community hospitals that receive patients to share the benefits with the tertiary hospital that sends patients depends on the size of member community hospitals and the cost of expanding the service capacity of community hospitals. Only when the size of community hospitals do not exceed a certain scale or their service capacity cost is not higher than a threshold or the subsidy budget is relatively shorter, the case that the government coordinates community hospitals to share their revenue is more conducive to the operation of hospital association referral service. Numerical experiments show that for the larger market demand, the size of community hospitals in the hospital association is the smaller or the service capacity cost is the lower, the government should coordinate community hospitals to share a higher revenue rate and provides a lower subsidy price to improve the patient referral rate, so as to improve the performance of each participant and reduce the waiting time. This paper proposes the contracts mechanisms for coordinating the effective operations of hospital association referral service, which provides theoretical decision support for the government to establish a scientific hierarchical healthcare delivery system. At the same time, this paper has a high application reference value to improve the overall efficiency of healthcare services.
Reference | Related Articles | Metrics
Mortality risk prediction for ICU patients based on time series feature representation and information fusion
XU Liangchen, GUO Chonghui
Systems Engineering - Theory & Practice    2022, 42 (10): 2815-2828.   DOI: 10.12011/SETP2022-0032
Abstract604)      PDF(pc) (1415KB)(240)       Save
Predicting the mortality risk of patients in the intensive care unit (ICU) is considered to be one of the most important clinical prediction tasks. Accurately predicting the mortality risk of ICU patients can provide information about the patient's condition and facilitate timely measures to intervene. At the same time, it contributes to the effective allocation of limited medical resources. The condition of ICU patients is unstable and requires close monitoring. A large amount of clinical data is collected, recorded and saved by relevant monitoring equipment, which provides an important reference for relevant clinical decision-making of ICU patients. This paper takes the 30-day mortality risk prediction of ICU patients as the research objective, based on the medical information mart for intensive care III database, constructs a mortality risk prediction model for ICU patients, and analyzes related influencing factors to support medical care clinical decision in practice. First extract relevant patient data and perform preprocessing, and use multiple statistics to represent time series features. The hybrid optimization algorithm SAGA based on the fusion of genetic algorithm and simulated annealing algorithm is used for feature selection of the corresponding base classifier, and finally uses the ensemble learning method to integrate the classifiers in bagging based on the idea of information fusion, and uses real data to verify the model. And based on the accuracy rate, F1 score and AUC three evaluation indicators, the model proposed in this article is compared with the classic mortality risk prediction model and showing better performance. Finally, based on information fusion, the important factors affecting the mortality risk of ICU patients are analyzed, and the central tendency statistics are more important, which provide reference for clinical decision-making.
Reference | Related Articles | Metrics
Optimization method of defense projects portfolio structure based on a mission-capability framework
LIN Mu, WANG Weiping, WANG Tao, ZHU Yifan, WANG Yanfeng, ZHOU Xin
Systems Engineering - Theory & Practice    2022, 42 (10): 2829-2839.   DOI: 10.12011/SETP2021-2622
Abstract325)      PDF(pc) (952KB)(168)       Save
Investment decision of national defense projects focuses on balancing the development of national defense forces and economic benefits, involving many essential elements, such as investment scale, investment fields, and underlying development prospects that affect the formation and development of national defense capabilities. This paper proposes a capability-based analytical framework for portfolio optimization based on the essential elements of investment decision-making in national defense projects. An optimization analysis framework with the hierarchical structure of project portfolio, capability, business, and mission is introduced, transforming the project portfolio optimization problem into a multi-objective optimization problem among capability, mission and cost. Then, guided by the analysis framework, a portfolio optimization decision-making method based on the NSGA-II algorithm is proposed. The optimization algorithm can be used as decision support for qualitative and quantitative decision-making of investment projects. Through case analysis, the applicability of this method is preliminarily proved.
Reference | Related Articles | Metrics