中国科学院数学与系统科学研究院期刊网

15 October 2025, Volume 45 Issue 10
    

  • Select all
    |
  • CAO Dong, ZHAO Jie, LI Wenwei, LAN Jingyu
    Journal of Systems Science and Mathematical Sciences. 2025, 45(10): 3021-3031. https://doi.org/10.12341/jssms240232
    Abstract & Html ( ) Download PDF ( )   Knowledge map   Save
    This paper uses the OP method and event study method to study the impact of blockchain technology application on enterprise total factor productivity and stock price, and analyzes whether the application of enterprise blockchain technology has effectively promoted the improvement of enterprise total factor productivity, or just created more “foam” for the company’s stock price? The main conclusion of this paper is that the application of blockchain technology mainly promotes the improvement of total factor productivity by reducing financing constraints, and has a greater impact on the improvement of total factor productivity for large enterprises and state-owned enterprises; In addition, after the application of blockchain technology in enterprise announcements, the company’s stock price level has significantly increased, meaning that the company can obtain higher stock premiums from blockchain technology based announcements.
  • GAO Wei, GENG Chen, WANG Dong, LI Xiuting
    Journal of Systems Science and Mathematical Sciences. 2025, 45(10): 3032-3046. https://doi.org/10.12341/jssms240299
    Abstract & Html ( ) Download PDF ( )   Knowledge map   Save
    This study constructs an ecological interconnected network consisting of 140 listed enterprises in the real estate industry, and utilizes network analysis methods and econometric models to investigate the evolutionary characteristics and cost effects of ecological interconnection among real estate enterprises. The findings indicate that the topological structure of the ecological interconnected network of real estate enterprises demonstrates consistent features aligned with China’s macroeconomic trends and industry development. This collaborative relationship based on ecological interconnected networks can generate significant cost effects, exhibiting strong heterogeneity due to variations in enterprise size and management capability. Policymakers should approach issues related to real estate industry development from a systemic and comprehensive perspective, mobilizing positive cooperation initiatives among industry enterprises in order to strive for collective industry advancement.
  • WEI Guanghe, YANG Chenghu, LI Xiaochao
    Journal of Systems Science and Mathematical Sciences. 2025, 45(10): 3047-3075. https://doi.org/10.12341/jssms240928
    Abstract & Html ( ) Download PDF ( )   Knowledge map   Save
    Blockchain technology effectively addresses trust issues in both product sales and waste recycling processes. However, its implementation increases the overall operational costs of the supply chain. How to balance the benefits and costs of blockchain investment has become a critical management challenge in low-carbon E-commerce closed-loop supply chains (E-CLSCs) under different sales models. To address this issue, this study investigates a low-carbon E-CLSC composed of a single manufacturer, an e-commerce platform, and an online retailer. Based on the manufacturer’s blockchain investment decision and the differences in sales models, the supply chain is categorized into four distinct types. A game-theoretic model is developed to explore the manufacturer’s profit-driven incentives for blockchain adoption and to reveal the underlying mechanisms through which key factors influence the investment decision. The research findings indicate that: 1) An increase in blockchain investment within an appropriate range can enhance the profits of all members as well as the overall system. Meanwhile, while an increase in the sales commission rate (i.e., the degree of price differentiation) within a specific range does not affect the profit growth of e-commerce platforms (or online retailers), it may reduce (or improve) the performance of other members and the overall system. From a waste recycling perspective, increased verification costs improve the effectiveness of waste product recycling when manufacturers delegate the process to e-commerce platforms. 2) When blockchain technology is implemented, manufacturers achieve higher profits under the agency selling model. However, when blockchain investment level, price differentiation, and sales commission rates are relatively low, the reselling model achieves higher profits for both low-carbon technology and e-commerce platforms, while the agency selling model benefits network retailers more, and vice versa. 3) Compared to the absence of blockchain technology, a low level of blockchain investment increases overall system profits across all sales model, but manufacturers earn higher profits in the agency selling model when blockchain investment is relatively low. The increase in low-carbon technology levels and profits for other stakeholders depends on a higher blockchain investment level. 4) Blockchain investment enhances profits for each member and the overall system only when low-carbon technology significantly impacts the demand discount factor or price differentiation, or when low-carbon technology investment costs are relatively low. If the sales commission rate or waste product verification fees are lower, the agency selling model yields higher profits for all members and the overall system.
  • MA Yanfang, WANG Yu, LI Zongmin, QIU Rui
    Journal of Systems Science and Mathematical Sciences. 2025, 45(10): 3076-3098. https://doi.org/10.12341/jssms23832
    Abstract & Html ( ) Download PDF ( )   Knowledge map   Save
    In real life, sometimes broken-down home appliances need to be shipped back to after-sales service center, and the home appliance after-sales service has problems such as long response time and high logistics operation cost. Considering real-time order information and dynamic vehicle information, an instant cycle scheduling system is designed, and an instant pickup routing optimization with heterogeneous fleet was formulated for home appliance after-sales service platform. Then, instant cyclic scheduling genetic algorithm is proposed to solve the model, in which initialization based on minimum wait time, roulette selection combined with elite strategy, best-cost route crossover, and inversion mutation are adopted. Based on the Solomon and Kilby benchmark instances, the performance of the ICSGA in solving advance order and real-time order was validated. Finally, heterogeneous fleet analysis, and sensitivity analysis for dynamic degrees and cut-off time of the platform are performed based on the Solomon instances adapted to some real-world situation. According to the results, management suggestions are provided for rational resource allocation and path planning of home appliance after-sales service enterprises.
  • SU Dongfeng, GUO Yi, PAN Yuxin
    Journal of Systems Science and Mathematical Sciences. 2025, 45(10): 3099-3110. https://doi.org/10.12341/jssms250320
    Abstract & Html ( ) Download PDF ( )   Knowledge map   Save
    To address the challenges of missing incremental information in time-series data and data distortion caused by impact disturbances in dynamic multi-attribute decision-making, this paper proposes a dynamic grey relation TOPSIS evaluation method based on global improved normalization with weakened buffer operators. First, a second-order buffer processing is applied to raw data using the average weakening buffer operator (AWBO) to mitigate disturbance impacts. Subsequently, a global improved normalization method is integrated to eliminate dimensional differences among indicators while preserving temporal incremental information, thereby constructing a standardized matrix. Further, grey relational analysis is fused with an enhanced TOPSIS approach by introducing a vertical distance orthogonal projection method to establish grey relation orthogonal projection closeness, comprehensively incorporating both indicator discrepancy and growth trends. Finally, a dynamic comprehensive evaluation is achieved using a “recent-priority” quadratic weighting strategy. The proposed method is validated through a case study on the scientific and technological innovation capabilities of universities in 10 eastern Chinese provinces and cities from 2019 to 2023. The results demonstrate its effectiveness in resolving data distortion and retaining incremental information in dynamic evaluations, providing theoretical support for multi-dimensional temporal decision-making problems.
  • LIU Zhifeng, ZHANG Qin, ZHANG Tingting
    Journal of Systems Science and Mathematical Sciences. 2025, 45(10): 3111-3134. https://doi.org/10.12341/jssms240211
    Abstract & Html ( ) Download PDF ( )   Knowledge map   Save
    This study approaches typhoon landfalls as exogenous climate risk events, designating the moment of landfall as the critical intervention point. Utilizing the difference-in-differences (DID) methodology, the research examines the influence of typhoon disasters on the stock returns of publicly traded companies in China, and assesses how financial risks propagate through supply chain networks triggered by typhoon disasters. To gain a more nuanced understanding of these effects, the paper engages in a detailed mechanism analysis by examining the intensity of digital transformation. The results suggest that typhoon disasters have a significant and detrimental impact on the stock returns of firms located in affected areas, with this effect rippling through to their suppliers and customers via the intricate web of supply chain connections. Moreover, the study uncovers a distinct asymmetry in the spillover effects between suppliers and customers. Specifically, the research highlights that the level of digital transformation is instrumental in alleviating the financial risks associated with typhoons and serves as a protective barrier against the adverse effects on stock returns. Finally, a comprehensive suite of robustness checks reinforces the validity and reliability of the study’s conclusions.
  • HE Zhifang, ZHONG Miaoqing
    Journal of Systems Science and Mathematical Sciences. 2025, 45(10): 3135-3156. https://doi.org/10.12341/jssms240301
    Abstract & Html ( ) Download PDF ( )   Knowledge map   Save
    In this paper, we investigate the dynamic relationship between climate policy uncertainty, global energy prices and stock prices using Granger causality test, time-varying parameter vector autoregressive (TVP-VAR) model and time-varying parameter vector autoregressive spillover index (TVP-VAR-DY) model. The results show that climate policy uncertainty is a Granger cause of global energy prices and stock prices, especially when a major climate policy event occurs, while there is a two-way Granger causality between global energy prices and stock prices. The results of the TVP-VAR model further show that the short-term impact of climate policy uncertainty on global energy prices is significantly positive while the medium- and long-term impacts turn from negative to positive, with the 2008 being the inflection point. Meanwhile, the impact of climate policy uncertainty on stock prices was generally negative until 2015, after which the impact manifests positive. Finally, based on the TVP-VAR-DY model, it is found that the total spillover effects of climate policy uncertainty, global energy prices and stock prices have obvious time-varying characteristics. Climate policy uncertainty is the spillover exporter, while global stock prices are the spillover receiver, and global energy prices have changed from the spillover receiver to the spillover exporter after 2008.
  • WAN Nana, FAN Jianchang, WU Xiaozhi, DU Juan
    Journal of Systems Science and Mathematical Sciences. 2025, 45(10): 3157-3170. https://doi.org/10.12341/jssms240323
    Abstract & Html ( ) Download PDF ( )   Knowledge map   Save
    This paper considers a two-stage retailer-led supply chain consisted of one supplier and one dominant retailer. Based on option contract and portfolio contract with options, this paper builds the multi-period dynamic game models. By using stochastic dynamic program, this paper analyzes the supplier’s multi-periodic optimal production policy and the retailer’s multi-periodic optimal ordering policy under two contracts, and provides an approximate algorithm to estimate the corresponding policy parameters. On this basis, this paper discusses the effect of two option contract formats on the performances of two members. The result suggests that option contract benefits the supplier, while portfolio contract with options benefits the retailer. Finally, the above conclusions are verified by an example analysis.
  • XIA Xuan, GONG Zaiwu
    Journal of Systems Science and Mathematical Sciences. 2025, 45(10): 3171-3192. https://doi.org/10.12341/jssms240402
    Abstract & Html ( ) Download PDF ( )   Knowledge map   Save
    In group decision-making, decision makers often demonstrate empathy towards others’ opinions due to the incompleteness of information in preference relations. This empathy effect is conducive to inspiring experts to provide higher-quality judgments. Addressing incomplete fuzzy preference relations with unknown weights, this paper proposes the ordinal regression method for incomplete preference relations considering empathy relations by integrating indirect preference information with low cognitive requirements. Firstly, based on the transformed empathy-induced indirect preference information, we construct the ordinal regression completion model and the conflicting information adjustment model. Then, by combining assessment information including empathy centrality, influence strength, and consensus measure, as well as indirect node information, the utility of each node is determined as the weight of the decision-maker through the construction of the ordinal regression model. Finally, consensus convergence is achieved through the minimum cost adjustment model. The proposed method considers the impact of empathetic network and indirect preference information on missing values and node utilities, which not only resolves logical conflicts caused by rough indirect information but also reduces the cost of consensus and improves the consistency and reliability of estimation results. Case analysis and comparative discussions demonstrate the effectiveness of the proposed method.
  • YUAN Pengcheng
    Journal of Systems Science and Mathematical Sciences. 2025, 45(10): 3193-3214. https://doi.org/10.12341/jssms23808
    Abstract & Html ( ) Download PDF ( )   Knowledge map   Save
    The process of optimizing ridepool matching decisions can be seen as a strategic game of decision-making among the ridepool management platform, drivers, and passengers. Based on this fundamental principle, this study introduces a personalized biding strategy (PBS: Personalized biding strategy) for ridepooling, incorporating it into the overall ridepool order optimization process to enhance the success rate of ridepooling. Initially, the study identifies two crucial factors that impact ridepool service quality: Detour distance and lateness duration. It presents passenger distance pricing functions and lateness penalty pricing functions based on these factors. Building upon this foundation, two models are developed: The dominant model of order optimization, which aims to maximize net profit, and the follower model for ridepool pricing (FMR), which aims to maximize actual travel utility. An optimized game model for personalized pricing and order planning considering service quality is constructed, taking into account service quality in personalized ridepool pricing and order planning. In this model, the ridepool management platform, as the dominant party, maximizes its profit by making decisions regarding order allocation and route execution. Subsequently, passengers, as followers, provide their desired ridepool prices based on the services offered by the platform’s order planning. A decomposition matching algorithm is proposed to solve this game model. The effectiveness of PBS in improving the ridepool success rate is validated through 56 different scenarios with 20 different parameter combinations. The results demonstrate that the PBS proposed in this study significantly improves the profitability of the ridepool platform, the overall utility of passengers, as well as the ridepool success rates for both vehicles and passengers, when compared to the average biding strategy (ABS) and the fixed biding strategy without considering service quality (FBS).
  • CHEN Ran, JIANG Wuyuan, YANG Jiayue
    Journal of Systems Science and Mathematical Sciences. 2025, 45(10): 3215-3227. https://doi.org/10.12341/jssms23536
    Abstract & Html ( ) Download PDF ( )   Knowledge map   Save
    This paper mainly studies the insurer’s robust optimal investment under the Black-Scholes model and a loss dependent premium principle. Assume that the claims process follows the Brownian motion with drift, the insurer is allowed to invest in a risk-free bond, a stock and a European call option. The price of stock is subject to Geometric Brownian Motion, and the insurer can buy proportional reinsurance from the reinsurer to diversify the investment risks. By solving the HJB equation, the analytic expressions of robust optimal reinsurance and investment strategies under the CARA utility are obtained, reinsurance and investment strategies of insurer under hedging conditions are also given. Finally, the influences of the model parameters on the optimal strategies are analyzed by numerical simulation.
  • LI Jiumin, XIA Dengfeng, FEI Weiyin, LI Guanjun
    Journal of Systems Science and Mathematical Sciences. 2025, 45(10): 3228-3244. https://doi.org/10.12341/jssms240289
    Abstract & Html ( ) Download PDF ( )   Knowledge map   Save
    This paper studies the reinsurance and investment game between an insurer and a reinsurer with relative performance concerns. We consider the joint interests of the insurer and the reinsurer for the reinsurance contract design. Namely, the insurer determines the claim risk sharing strategy, the reinsurer determines the reinsurance price, and they codetermine the final reinsurance premium. To increase their respective wealth, the insurer and the reinsurer can invest in the same risk-free asset and risky asset which follows the constant elasticity of variance (CEV) model. We quantify the competition between the insurer and the reinsurer through their relative performances. Both of them aim at maximizing the expected value of their terminal relative wealth while minimizing its variance. By using the stochastic optimal control technique, we formulate and solve extended Hamilton-Jacobi-Bellman (HJB) equations under the Stackelberg game framework. And the optimal reinsurance contract as well as the optimal time-consistent investment strategy are derived analytically. Finally, numerical simulations show that with relative performance concerns, the insurer would probably spend less on reinsurance, the reinsurer tends to lower the reinsurance price, and the final reinsurance premium decreases. Besides, both insurer and reinsurer would invest more in risky asset with relative performance concerns.
  • WANG Yufang, WANG Nan, ZHANG Shuhua
    Journal of Systems Science and Mathematical Sciences. 2025, 45(10): 3245-3266. https://doi.org/10.12341/jssms240059
    Abstract & Html ( ) Download PDF ( )   Knowledge map   Save
    To solve the problem of instability and imprecision of carbon price prediction with single information source, single decomposition technology and single prediction method, a hybrid prediction model of carbon price based on multi-source data feature and multi-scale analysis is proposed, called CPS-MEMD-SVR-MLR. 1) Multi-source data analysis: This effectively integrates historical carbon trading prices related to carbon prices, macroeconomic development levels, fossil energy prices, exchange rates, and social media sentiment data based on news text information; 2) Multi-scale analysis: This uses multiple empirical mode decomposition technology (MEMD) to decompose multi-source data into prediction features under different modes; 3) Hybrid prediction analysis: This uses fuzzy entropy theory to orderly integrate econometric model and machine learning models, and then integrates the predicted values of each mode into the final result. This paper takes the carbon price of the European Union (EU) from February 11, 2015 to February 27, 2023 as a case study. Based on seven scenarios and DM tests, the results show that: 1) The prediction accuracy of the hybrid model proposed in this paper is better than other comparison models; 2) Social media sentiment can improve the prediction accuracy of carbon price, and it is better than the single factor prediction; 3) The introduction of MEMD decomposition can significantly improve the prediction accuracy of carbon price.
  • WANG Shuying, MEI Wenjuan, MA Rui
    Journal of Systems Science and Mathematical Sciences. 2025, 45(10): 3267-3278. https://doi.org/10.12341/jssms23685
    Abstract & Html ( ) Download PDF ( )   Knowledge map   Save
    In practical research, the data may come from different distributions, and there are some limitations when using a single distribution model to fit the data. In order to overcome this problem, the mixed model can better adapt to complex data types. Exponential distribution and Rayleigh distribution are important life distributions in reliability analysis, and there are few related mixed models in the context of censored data. In this paper, a mixed model of two-parameter exponential distribution and two-parameter Rayleigh distribution is proposed, and the EM algorithm is used to estimate the parameters of the mixed model with right censored data. Finally, the model is applied to the actual data, and the goodness of fit test is carried out to verify that the proposed model is suitable, which further shows that the model can better adapt to the complex data characteristics and has certain practical significance.
  • PAN Yingli, WANG Haoyu, HUANG Yijing, XU Caixu, HUANG He
    Journal of Systems Science and Mathematical Sciences. 2025, 45(10): 3279-3298. https://doi.org/10.12341/jssms240153
    Abstract & Html ( ) Download PDF ( )   Knowledge map   Save
    This paper, set against the backdrop of big data, is based on a quantile regression model with covariates missing at random. It adopts the concept of distributed storage, randomly storing data across different machines. By constructing a communication-efficient surrogate loss function for the global loss function, the global optimization problem is transformed into a local optimization problem. A Proximal ADMM algorithm is designed to iteratively solve for the optimal estimator. This paper addresses the challenges of data storage and the high cost of communication between machines in quantile regression models with missing covariates. Theoretical research shows that, under certain regular conditions, the proposed distributed estimator is consistent and asymptotically normal. Numerical analysis demonstrates that, with a limited number of communications between the master and slave machines, the estimation error of the proposed distributed optimization method decreases and converges to the estimation error obtained by the globally optimal Oracle method. Moreover, it yields smaller estimation errors compared to the average-based OneShot method and weighted least squares regression.
  • LIN Jinguan, REN Yang, WANG Jiangyan
    Journal of Systems Science and Mathematical Sciences. 2025, 45(10): 3299-3317. https://doi.org/10.12341/jssms240127
    Abstract & Html ( ) Download PDF ( )   Knowledge map   Save
    Covariance estimation poses a crucial challenge in the analysis of high dimensional data, which in turn is prone to the two phenomena of heavy-tailed distributions, small samples, and in these cases, the traditional estimation methods (e.g., sample covariance matrix) prove inadequate for such heavy-tailed data, given their lack of accuracy. In cases where heavy-tailed high dimensional data represented as tensors (multi-dimensional arrays), harnessing the tensor structure is a good choice for achieving dimensionality reduction. To this end, this paper proposes novel structured regularization methods for estimating the covariance of heavy-tailed tensor-valued data. In this paper, the heavy-tailed tensor data are first truncated, then the truncated sample covariance matrix is computed, and the CP decomposition will be applied to find an approximation in the form of Kronecker product of multiple matrices of the truncated sample covariance matrix, and finally imposes a banded or tapering structure for each of the small matrices obtained by the decomposition. Simulation results show that the proposed estimators have excellent performance for different degree of heavy tailing and different sample sizes. Anomalous temperature datasets with heavy-tailed distributions is analysed using the estimation method proposed in this paper.
  • YU Lu, LI Ting, LóPEZ-CARR David, HU Guihua, WU Di, QI Li
    Journal of Systems Science and Mathematical Sciences. 2025, 45(10): 3318-3333. https://doi.org/10.12341/jssms240158
    Abstract & Html ( ) Download PDF ( )   Knowledge map   Save
    This paper aims to introduce the demographic analysis model of the American and establish a demographic analysis model, which is suitable for application in China. Literature interpretation and probability modeling methods is used to study demographic analysis models and related issues. The results show that the research of demographic analysis model in China is still in its infancy, and the establishment of demographic analysis model in China can not copy the American demographic analysis model, but should be improved and innovated in combination with the actual situation in China; The establishment of demographic analysis models can be divided into two levels. One is to establish a demographic analysis model based on comprehensive data, without calculating sampling variance, the other is to establish a demographic analysis model based on sampling survey data, which needs to calculate its sampling variance by jack knife method approximately; The advantage of the demographic analysis model is that it fully utilizes population administrative record data such as birth, death, and international migration, saves data collection costs, and is independent of the population census, the disadvantage is that the estimation results are uncertain, especially with significant differences in the estimated results of international net migration count. The demographic analysis model is expected to be applied to estimate the net error of China’s 2030 population census, creating a precedent for China to use demographic analysis model in this field.
  • JIANG Xiaoting, SUN Jinxuan, GUO Baocai
    Journal of Systems Science and Mathematical Sciences. 2025, 45(10): 3334-3359. https://doi.org/10.12341/jssms23916
    Abstract & Html ( ) Download PDF ( )   Knowledge map   Save
    The variance of the quality characteristic shows the variation of a product or process. If the variance increases, it indicates that the product or process has deteriorated; Otherwise, it indicates the product or process has improved. It is crucial to design a two-sided tolerance interval for the population of sample variances in order to reasonably evaluate the performance of a product or process. For the traditional “equal-tailed” β-expectation tolerance interval for sample variances, when the number of subgroups or subgroup size is particularly small, there are two major problems: 1) Ignoring the actual coverage variation leads to significant differences among practitioners; 2) The accuracy level of the tolerance interval is too low, that is, the minimum number of subgroups required to achieve the desired accuracy level is too large. Small sample size often occurs due to limited time or high cost in the field of quality control; thus, this paper focuses on improving the accuracy level and designs new β-expectation tolerance intervals for the population of sample variances based on the actual coverage variance minimization method and the actual coverage centralization method. This paper also introduces a Bayesian method to design the corresponding tolerance intervals, considering that the practitioner usually has certain prior information due to historical data or experience. The performance of the tolerance intervals is evaluated and compared using the accuracy level and the minimum number of subgroups to achieve the desired accuracy level. The results show: 1) The proposed tolerance intervals outperform the traditional “equal-tailed” one; 2) The performance of the Bayesian tolerance intervals is better than the corresponding frequentist ones. Finally, a real example is used to demonstrate the design and superiority of the proposed β-expectation tolerance intervals.
  • LI Angyan, ZHAO Chenyan, LU Lizheng
    Journal of Systems Science and Mathematical Sciences. 2025, 45(10): 3360-3370. https://doi.org/10.12341/jssms240532
    Abstract & Html ( ) Download PDF ( )   Knowledge map   Save
    To interpolate the specified Frenet frame, curvature and torsion, a method is proposed for the construction and shape optimization of spatial quintic F3 interpolating curves. F3 continuity of spatial curves is a special k-th order Frenet frame continuity and ensures the satisfaction of G2 continuity and torsion interpolation. Firstly, a quintic Bézier curve interpolating the given F3 data is constructed, whose control points are expressed with two parameters denoting the lengths of the curve’s end tangents. Then, the optimal parameter values are determined by minimizing a quadratic energy function. Finally, by defining the objective function as the integral of a weighted sum of squared curvature and torsion, another better optimization method is proposed. Compared to the previous G2 interpolation scheme, the new methods can generate curve shapes with more satisfactory curvature and torsion profiles, although using a stricter continuity constraint.
  • GU Hengyang, DU Xuewu
    Journal of Systems Science and Mathematical Sciences. 2025, 45(10): 3371-3384. https://doi.org/10.12341/jssms240367
    Abstract & Html ( ) Download PDF ( )   Knowledge map   Save
    Among traditional gradient-like methods for solving unconstrained optimization problems, conjugate gradient method has the advantages of small storage requirement, simple iterative form and fast speed of computation. Barzilai-Borwein (BB) gradient methods are a class of improved algorithms for steepest descent method. They have good theoretical convergence and can avoid the zigzag phenomenon of steepest descent method. Spectral conjugate gradient methods are a class of conjugate gradient methods with good numerical performance and they use one of stepsizes in BB gradient methods as the spectral parameter. In this paper, we choose the parameter in a family of Dai-Kou (DK) conjugate gradient methods as the negtive of the reciprocal of another stepsize in the BB gradient methods. Furthermore, by combining Fletcher-Reeves (FR) conjugate gradient method which has good theoretical convergence with a variant of Polak-Ribière-Polyak (PRP) conjugate gradient method which has good computational performance, we present a class of hybrid truncated conjugate gradient methods with a convex combination form. In order to improve the numerical performance of this class of methods, we present a class of hybrid truncated spectral conjugate gradient methods with a restart step by combining a restart strategy and the idea of spectral conjugate gradient method. The choice of the spectral parameter guarantees that the methods in this paper possess the sufficient descent property without relying on any line search. Numerical experiment results show that the algorithm given in this paper has better numerical performance than the DK, DK+, PRP and a hybrid Dai-Yuan (HDY) conjugate gradient algorithms. Finally, we verify again the effectiveness of our algorithm by applying them for solving image restoration problems.