No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Assessing project portfolio risk via an enhanced GA-BPNN combined with PCA
Abstract
Assessing project portfolio risk (PPR) is essential for organizations to grasp the overall risk levels of project portfolios (PPs) and realize PPR mitigation. However, current research is inadequate to effectively assess PPR, which brings challenges to managing PPR. In this context, the purpose of this study is to develop a PPR assessment model via an enhanced backpropagation neural network (BPNN). First, PPR assessment criteria considering project interdependencies are determined. Second, fuzzy logic is used to obtain original data for assessment criteria. Principal component analysis (PCA) is then employed to reduce the dimensionality of assessment criteria and derive the input and output of BPNN. Third, an improved genetic algorithm (IGA) is designed to optimize the initial weights and thresholds of BPNN. On this basis, the PCA-IGA-BPNN assessment model is constructed, followed by training and testing, possessing a test accuracy of 98.6%. Finally, comparison experiments are conducted from both internal and external perspectives. For internal comparison, the proposed model yields less mean absolute percentage error (MAPE), mean square error (MSE), and root mean square error (RMSE) than PCA-GA-BPNN, IGA-BPNN, PCA-BPNN and BPNN and offers the largest convergence speed (). As for external comparison, the presented model produces lower MAPE, MSE, and RMSE than Decision Tree (DT), Support Vector Machine (SVM), Random Forest (RF), and K-Nearest Neighbor (KNN) and has the largest coefficient of determination (). Results indicate that the established model performs more satisfactorily in assessing PPR. This research enriches PPR assessment methods and provides managers with a useful tool to evaluate PPR.
... These ESRs have caused significant losses at various stages of PP [3], and managing ESRs is an essential aspect of achieving project portfolio management (PPM) objectives [4]. Any PPM approach that neglects ESRs may result in an unbalanced PP, highlighting the important role of managing these risks in PP [5]. ESR management is a part of risk management, including risk identification, assessment and response [1]. ...
To improve the effectiveness of external stakeholder risks (ESRs) management in project portfolios (PPs), a portfolio-wide risk response approach is required. However, current research is inadequate to effectively identify response strategies for ESRs, which brings challenges to managing ESRs in PPs. In this context, the purpose of this study is to select an appropriate combination of response strategies for ESRs by considering interactions among ESRs, projects, and response strategies in the PP. A Bayesian influence diagram (BID) coupled with a multi-objective optimization model is deemed suitable for this context. Firstly, a probability-sensitivity matrix is established to determine the key ESRs. Then, a BID is constructed to calculate the expected values of different combinations of response strategies. Finally, integrating stakeholder satisfaction and strategy cost, an optimization model for risk response strategy selection is established to obtain candidate combinations. By combining expected values and candidate combinations, the optimal strategy combination is selected. The proposed model comprehensively considers and evaluates the interactions between risks, projects, and risk responses. This enhances the desirability of expected outcomes and reduces project execution costs.
... Hossain and Wu employed SNA to analyze the relationship between network centrality and project coordination efciency [27], while Son and Rojas used SNA to examine network efciency, network stability, and network cohesion in engineering project construction [28]. However, existing articles on social network analysis primarily focus on analyzing the impact of network metrics on other factors [29], with limited emphasis on analyzing the behavior and decision-making of individual actors within the network. ...
The indexing work of complex products and systems (CoPS) is to determine a comprehensive set of indicators to ensure seamless integration of CoPS modules. Despite its pivotal role as a pioneering phase in CoPS development, indexing work has garnered relatively scant research attention. This paper explores CoPS indexing work through a network game model that considers a main manufacturer and several suppliers. The primary aim is to discern the key influencing factors affecting CoPS indexing work and elucidate the influence mechanism. Several interesting conclusions have been drawn: (1) subjective, environmental, and structural factors as the three key aspects influencing the CoPS indexing work network; (2) Subjective factors directly influence the agencies’ selection of the optimal effort level for collaborative work, whereas environmental and structural factors indirectly impact their choice by affecting their network centrality; (3) the environmental factors within the indexing work network dictate the rate at which interagent interactions diminish with distance. To demonstrate and validate the research’s findings, an examination of the indexing process of the China Lanxin High-speed Railway is conducted. This study offers new insights into CoPS indexing work, providing both theoretical references and practical suggestions for project teams to improve collaborative efficiency.
... Zhang et al. [83] proposed the weight calculation method of Group Analytic Hierarchy Process-Principal Component Analysis to rank the critical construction risk factors. Most studies use PCA as a data preprocessing tool [84,85]. Nevertheless, it also has some disadvantages. ...
Machine learning, a key thruster of Construction 4.0, has seen exponential publication growth in the last ten years. Many studies have identified ML as the future, but few have critically examined the applications and limitations of various algorithms in construction management. Therefore, this article comprehensively reviewed the top 100 articles from 2018 to 2023 about ML algorithms applied in construction risk management, provided their strengths and limitations, and identified areas for improvement. The study found that integrating various data sources, including historical project data, environmental factors, and stakeholder information, has become a common trend in construction risk. However, the challenges associated with the need for extensive and high-quality datasets, models’ interpretability, and construction projects’ dynamic nature pose significant barriers. The recommendations presented in this paper can facilitate interdisciplinary collaboration between traditional construction and machine learning, thereby enhancing the development of specialized algorithms for real-world projects.
... Generally, multiple projects within PP share some same resources, including human, equipment, materials, capital, information, technology, and other resources during overlapping execution [44]. When a project is interrupted or delayed, allocating SRs preferably to these projects could reduce the negative impact on the project schedule. ...
To reasonably allocate shared resources (SRs) in project portfolios (PP) and realize PP's maximal benefits, this research proposes a hybrid SRs allocation model based on system dynamics (SD) and discrete event simulation (DES). Unlike previous resource allocation methods that solely consider project level benefits, the proposed model simultaneously considers the benefits at both project and portfolio levels. Specifically, starting from the project activity level, we first construct a project schedule evaluation sub model using DES; Second, a synergy benefit evaluation sub model at portfolio level is constructed by considering both the positive and negative benefits brought by SRs allocation using SD; Finally, by clarifying the connection between the two sub models, a hybrid simulation model for SRs allocation is obtained. A case study is used to demonstrate the practicality of the hybrid model. Sensitivity analysis is then conducted to examine how the model output changes due to variations in parameters. The proposed model is also compared with DES model to demonstrate the superiority of the hybridization. Simulation results reveal that the proposed model can systematically integrate the effects of SRs allocation at both project and portfolio levels, which provides PP managers a tool to enhance SRs allocation performance.
... Principal components are another name for eigenvectors. They are initially subject to standard normalization because of the different feature domains [15]. ...
Nowadays, the diagnosis of breast cancer (DBC) helps doctors make early detection of breast cancer into non-cancerous (benign B) and cancerous (malignant M). Therefore, using machine learning (ML) algorithms is a solution to diagnosing and predicting symptoms related to DBC. The increased computational complexity, data size, overfitting, and longer training times harm early diagnosis accuracy. In this paper, propose a dimensionality reduction model integrating PCA and KNN for early breast cancer detection. which is used to diagnose and predict breast cancer (DPBC) based on reduced data size by selecting the best features that capture most of the variance in the data. The performance of the proposed model is evaluated with indices such as accuracy, precision, and f-score. Results for the DPBC model were obtained by using the Breast Cancer Wisconsin medical datasets (BCW).
... Currently, PPSS problems with a set of constraints have been identified as NP-hard in the studies, and the objective function in the proposed mathematical model has the characteristic of non-linearity. It is of great benefit to employ meta-heuristics algorithms, such as genetic algorithm (GA), colony optimization (ACO) and SAA, to solve the PPSS problem due to their powerful ability of handing optimization problem (Bai et al., 2023b). Among those metaheuristic algorithms, SAA was characterized by easy implementation, fast convergence and jumping out of the local solution for global search about handling many NP-hard optimization problems effectively (Bhuvaneswari and Subramanian, 2005;Kolahan et al., 2011). ...
Purpose
Selecting and scheduling optimal project portfolio simultaneously is a complex decision-making problem faced by organizations to realize the strategy. However, dynamic synergy relationships among projects complicate this problem. This study aims at constructing a project portfolio selection and scheduling (PPSS) model while quantifying the dynamic synergetic effects to provide decision support for managing PPSS problems.
Design/methodology/approach
This study develops a mathematical model for PPSS with the objective of maximal project portfolio benefits (PPBs). To make the results align with the strategy, comprehensive PPBs are divided into financial and non-financial aspects based on the balanced scorecard. Then, synergy benefits evolve dynamically in the time horizon, and system dynamics is employed to quantify them. Lastly, a case example is conducted to verify the applicability of the proposed model.
Findings
The proposed model is an applicable model for PPSS while incorporating dynamic synergy. It can help project managers obtain the results that which project should be selected and when it should start while achieving optimal PPBs.
Originality/value
This study complements prior PPSS research in two aspects. First, financial and non-financial PPBs are designed as new criteria for PPSS, making the results follow the strategy. Second, this study illuminates the dynamic characteristic of synergy and quantifies the synergetic effect. The proposed model provides insights into managing a PPSS effectively.
... As previously stated, ANN is a kind of artificial intelligence that is similar to architectural structure and functions like brain cells or the human nervous system. It can learn from experience and provide reliable answers to complex problems [30,31]. As an efficient methodology, ANN has been extensively applied in the field of service provider selection. ...
During the project life cycle, selecting the appropriate service provider portfolio (SPP) is essential to guaranteeing the successful implementation of manufacturing tasks. However, the existence of the synergy effect among service providers poses a challenge for decision makers in selecting the most suitable SPP. To effectively address this challenge, this study constructs a novel service provider portfolio selection (SPPS) model across the project life cycle, considering the synergy effect. The model is based on the integration of a radial basis function neural network (RBFNN), the technique for order preference by similarity to ideal solution (TOPSIS), and the entropy method (EM). First, the evaluation criteria for service provider selection are defined, followed by the identification of alternative service providers and feasible SPPs based on project life cycle division. Subsequently, a quantitative analysis of the synergy effect among service providers within the same stage, as well as between different stages, throughout the project life cycle, is carried out. This analysis helps to determine the input variables and expected output variables that will be utilized in the model. Additionally, the feasibility and applicability of the proposed model are illustrated through an example. Finally, a comparison between the proposed hybrid model and the BPNN is conducted to validate the model's accuracy and efficiency. This study contributes to making sound decisions in the SPPS process from a project life cycle perspective.
... Furthermore, the types of PPRs and the interaction relationships between them will change dynamically along with the PP implementation. Specifically, different types of PPRs may arise at different times during the PP implementation, and the interactions between different types of PPRs are diverse [9]. However, existing studies appear to focus on assessing PPRs from a static perspective [10,11], ignoring the dynamic changes of PPRs themselves and their interactions with time. ...
Due to dynamic changes in both internal and external environments, the risk evolution of the project portfolio (PP) becomes extremely complicated, thereby increasing the difficulties of effective risk response. In particular, the real-time influence of risk interactions on the evolution of project portfolio risk (PPR) often goes unnoticed. Meanwhile, risk contagiousness is completely ignored in risk response. To tackle this challenge, this study proposes a PPR evolution and response (PPRER) model by improving the Barrat–Barthelemy–Vespignani (BBV) model and by introducing the evolutionary dynamics method into the PPR response research. The feasibility and applicability of the proposed model are verified through a numerical illustration. Computational results demonstrate that the proposed model can simulate the evolution process of PPRs under the influence of their interactions and give a snapshot of the real-time interactive relationship between PPRs. Based on the obtained results, decision-makers can take effective risk responses by identifying critical strategy intrusion nodes at any time in the evolution process.
... With = 0, < . Given the initial risk known within a collection of PPP projects, the anticipated loss of the initial risk , denoted as = ( , ) = × [27]. In scenarios where the control of the initial risk yield secondary risks, each secondary risk carries an expected loss of = , = × [19]. ...
Existing research on risk sharing management often ignores the adverse consequences of secondary risks. This study addresses secondary risks that emerge from the implementation of specific risk mitigation measures. Addressing the limitations of existing research that overlooks secondary risks and exploring the impact of secondary risks on the outcome of risk sharing, this paper integrates secondary risks into a game model that examines risk sharing between the public and private sectors in EPC+PPP projects. Utilizing a risk-control benefit model, an evolutionary game model is established to determine the evolutionary stability strategy under various conditions. This encompasses factors such as project income distribution, risk control capability, imbalanced status, and risk compensation. The findings indicate that secondary risks impact the risk-sharing strategies of both parties. Furthermore, a stronger risk-control ability correlates with a greater inclination toward risk-taking. The public sector can motivate risk sharing for the private sector through risk compensation, with a discernible lower limit. Ultimately, risk sharing becomes an active choice for both public and private entities when the risk cost is below the difference in returns between risks.
To promote the dynamic risk management of construction project portfolios (CPPs) and improve the effect of risk response, a simulation-optimization model is proposed. The model integrates System Dynamics and optimization to dynamically select risk response strategies (RRSs) and facilitate more refined resource allocation. Specifically, a System Dynamics sub-model simulates the risk level by capturing one-stage and 2 cross-stage risk interactions. Second, develop an optimization sub-model to select RRSs with resource allocation, which allows for partially allocated resources for RRSs. Afterward, integrate them into a simulation-optimization model. The closed-loop information flow between the two sub-models enables the assessment of the initial risk at a specific stage, the selection of optimal RRSs, and the updating of the dynamic impact of the risks of that stage on later stages. Finally, the proposed model is validated using a numerical example. The findings accentuate the significance of considering cross-stage risk interactions and optimizing resource allocation in the risk response of CPPs, and validate the feasibility and superiority of the model for solving the dynamic selection of RRSs.
Purpose
Prefabricated buildings (PBs) have proven to effectively mitigate carbon emissions in the construction industry. Existing studies have analyzed the environmental performance of PBs considering the shift in construction methods, ignoring the emissions abatement effects of the low-carbon practices adopted by participants in the prefabricated building supply chain (PBSC). Thus, it is challenging to exploit the environmental advantages of PBs. To further reveal the carbon reduction potential of PBs and assist participants in making low-carbon practice strategy decisions, this paper constructs a system dynamics (SD) model to explore the performance of PBSC in low-carbon practices.
Design/methodology/approach
This study adopts the SD approach to integrate the complex dynamic relationship between variables and explicitly considers the environmental and economic impacts of PBSC to explore the carbon emission reduction effects of low-carbon practices by enterprises under environmental policies from the supply chain perspective.
Findings
Results show that with the advance of prefabrication level, the carbon emissions from production and transportation processes increase, and the total carbon emissions of PBSC show an upward trend. Low-carbon practices of rational transportation route planning and carbon-reduction energy investment can effectively reduce carbon emissions with negative economic impacts on transportation enterprises. The application of sustainable materials in low-carbon practices is both economically and environmentally friendly. In addition, carbon tax does not always promote the implementation of low-carbon practices, and the improvement of enterprises' environmental awareness can further strengthen the effect of low-carbon practices.
Originality/value
This study dynamically assesses the carbon reduction effects of low-carbon practices in PBSC, informing the low-carbon decision-making of participants in building construction projects and guiding the government to formulate environmental policies.
Purpose
Assessing project criticality in a project portfolio (PP) is of great practical significance to improve robustness from damage. While project criticality assessment has increased diversity in approaches, the understanding of vulnerable project impacts is still limited. To promote a better understanding of assessing project criticality, a vulnerability measurement model is constructed.
Design/methodology/approach
First, integrating the tasks, projects and corresponding relationships among them, a project portfolio network (PPN) is constructed. Second, the project's vulnerability is measured by combining the topological structure and functional attributes. Third, project criticality is assessed by the vulnerability measurement results. Lastly, the proposed model is applied in a numerical example to illustrate its suitability and effectiveness.
Findings
For academia, this study provides a novel perspective on project vulnerability measurement and expands project criticality assessment tools. For practitioners, the straightforward model provides an effective tool for assessing project criticality and contributes to enhancing project portfolio management (PPM).
Originality/value
The impact of the task on the project is considered in this study. Topological structure and functional attributes are also integrated for measuring project vulnerability due to the impact of random attacks in an uncertain environment, providing a new perspective on the requirements of project criticality assessment and the measurement of project vulnerability.
This paper proposes an efficient federated distillation learning system (EFDLS) for multi-task time series classification (TSC). EFDLS consists of a central server and multiple mobile users, where different users may run different TSC tasks. EFDLS has two novel components: a feature-based student-teacher (FBST) framework and a distance-based weights matching (DBWM) scheme. For each user, the FBST framework transfers knowledge from its teacher’s hidden layers to its student’s hidden layers via knowledge distillation, where the teacher and student have identical network structures. For each connected user, its student model’s hidden layers’ weights are uploaded to the EFDLS server periodically. The DBWM scheme is deployed on the server, with the least square distance used to measure the similarity between the weights of two given models. This scheme finds a partner for each connected user such that the user’s and its partner’s weights are the closest among all the weights uploaded. The server exchanges and sends back the user’s and its partner’s weights to these two users which then load the received weights to their teachers’ hidden layers. Experimental results show that compared with a number of state-of-the-art federated learning algorithms, our proposed EFDLS wins 20 out of 44 standard UCR2018 datasets and achieves the highest mean accuracy (70.14%) on these datasets. In particular, compared with a single-task Baseline, EFDLS obtains 32/4/8 regarding ’win’/’tie’/’lose’ and results in an improvement of approximately 4% in terms of mean accuracy.
With the continuous development of my country’s economy and society, how to effectively evaluate the technical innovation performance of agricultural enterprises has become the focus of research. This paper firstly processes and analyzes the technical innovation performance data of agricultural enterprises, and then processes and converts the technical innovation performance data of agricultural enterprises; then, through the analysis of the technical innovation performance data of agricultural enterprises, the key characteristics of the technical innovation performance of agricultural enterprises are excavated. Finally, a performance evaluation model based on agricultural enterprise technology innovation is proposed, and the validity of the model is verified with examples.
The construction of a green financial system can promote economic transformation and development, scientifically and effectively evaluate the green environmental protection level of finance in different cities, and provide a strong impetus for its future promotion. This paper introduces a BP neural network into the evaluation system of urban green finance progress and optimizes the model through GA. According to the constructed evaluation index system, this paper makes empirical analysis of the experimental city. The optimization results show that GA can improve the training error range of the BP neural network and increase the stability of evaluation model performance and the accuracy of evaluation results. Through empirical analysis, it is concluded that the development status of the urban economy, the mode and efficiency of capital distribution, and the degree of support for the environmental protection industry in terms of policy and capital will have a great impact on the development of its environmental protection finance. The support attitude of the local government and society in energy conservation, environmental protection, and green city construction provides a broad development space for green finance to a great extent. Cities should broaden the development channels of green finance and build a sound and scientific green financial service system.
In IoT applications, energy supply, especially wireless power transfer (WPT), has attracted the most attention in the relevant literature. In this paper, we present a new approach to laser irradiance solar cell panels and predicting energy transfer efficiency. From the previous experimental datasets, it has been discovered that in the laser charging (LC) process, temperature has a great impact on the efficiency, which is highly correlated with the laser intensity. Then, based on artificial neural network (ANN), we set the above temperature and laser intensity as inputs, and the efficiency as output through back propagation (BP) algorithm, and use neural network and BP to train and modify the network parameters to approach the real efficiency value. We also propose the genetic algorithm (GA) to optimize the learning rate of the BP, which achieved slightly superior results. The results of the experiment indicate that the prediction method reaches a high accuracy of about 99.4%. The research results in this paper provide an improved solution for the LC application, particularly the energy supply of IoT devices or small electronic devices through WPT.
An adaptive genetic algorithm based on collision detection (AGACD) is proposed to solve the problems of the basic genetic algorithm in the field of path planning, such as low convergence path quality, many iterations required for convergence, and easily falling into the local optimal solution. First, this paper introduces the Delphi weight method to evaluate the weight of path length, path smoothness, and path safety in the fitness function, and a collision detection method is proposed to detect whether the planned path collides with obstacles. Then, the population initialization process is improved to reduce the program running time. After comprehensively considering the population diversity and the number of algorithm iterations, the traditional crossover operator and mutation operator are improved, and the adaptive crossover operator and adaptive mutation operator are proposed to avoid the local optimal solution. Finally, an optimization operator is proposed to improve the quality of convergent individuals through the second optimization of convergent individuals. The simulation results show that the adaptive genetic algorithm based on collision detection is not only suitable for simulation maps with various sizes and obstacle distributions but also has excellent performance, such as greatly reducing the running time of the algorithm program, and the adaptive genetic algorithm based on collision detection can effectively solve the problems of the basic genetic algorithm.
Time series data usually contains local and global patterns. Most of the existing feature networks focus on local features rather than the relationships among them. The latter is also essential, yet more difficult to explore because it is challenging to obtain sufficient representations using a feature network. To this end, we propose a novel robust temporal feature network (RTFN) for feature extraction in time series classification, containing a temporal feature network (TFN) and a long short-term memory (LSTM)-based attention network (LSTMaN). TFN is a residual structure with multiple convolutional layers, and functions as a local-feature extraction network to mine sufficient local features from data. LSTMaN is composed of two identical layers, where attention and LSTM networks are hybridized. This network acts as a relation extraction network to discover the intrinsic relationships among the features extracted from different data positions. In experiments, we embed the RTFN into supervised and unsupervised structures as a feature extractor and encoder, respectively. The results show that the RTFN-based structures achieve excellent supervised and unsupervised performances on a large number of UCR2018 and UEA2018 datasets.
Economic early warning is the recognition and judgment of the state of economic operation, and its research results directly affect the rational formulation of macro-control policies. However, the traditional early warning methods are mainly based on expert experience or simple statistical model, which are difficult to reflect the nature of highly nonlinear economic system and can not meet the objective requirements of macroeconomic early warning. Based on the above background, the purpose of this study is to design an economic early warning system based on improved genetic and BP hybrid algorithm and neural network. Based on the overview of macroeconomic early warning at home and abroad, this study expounds the design of early warning index system, early warning model, establishment of early warning system and other issues in the macroeconomic early warning theoretical system; deeply analyses the theoretical methods of BP neural network and adaptive mutation genetic algorithm, and discusses the feasibility of realizing macroeconomic early warning by BP neural network and adaptive mutation genetic algorithm, The improved genetic and BP hybrid algorithm and neural network economic early warning model are established. Finally, the experimental results show that the correlation coefficient between the composite index and the comprehensive early warning is 0.89, and the delay number is 0, which shows that the early warning index obtained by the early warning system can accurately reflect the actual economic fluctuations. The results show that the improved genetic and BP hybrid algorithm and neural network economic early warning system are effective, feasible and have good accuracy.
The emergency management of chemical accidents plays an important role in preventing the expansion of chemical accidents. In recent years, the evaluation and research of emergency management of chemical accidents has attracted the attention of many scholars. However, as an important part of emergency management, the professional rescue team of chemicals has few evaluation models for their capabilities. In this study, an emergency rescue capability assessment model based on the PCA-BP neural network is proposed. Firstly, the construction status of 11 emergency rescue teams for chemical accidents in Shanghai is analyzed, and an index system for evaluating the capabilities of emergency rescue teams for chemicals is established. Secondly, the principal component analysis (PCA) is used to perform dimension reduction and indicators’ weight acquisition on the original index system to achieve an effective evaluation of the capabilities of 11 rescue teams. Finally, the indicators after dimensionality reduction are used as the input neurons of the backpropagation (BP) neural network, the characteristic data of eight rescue teams are used as the training set, and the comprehensive scores of three rescue teams are used for verifying the generalization ability of the evaluation model. The result shows that the proposed evaluation model based on the PCA-BP neural network can effectively evaluate the rescue capability of the emergency rescue teams for chemical accidents and provide a new idea for emergency rescue capability assessment.
To further explore the influence path of internet finance on the risk prevention and management of commercial banks, the backpropagation neural network optimization algorithm was used to predict the risk value and the change of the risk level of commercial banks under the background of internet environment was empirically studied and analyzed. The results showed that the maximum size of genetic algebra and the number of individuals significantly impacted the algorithm’s optimization performance when the genetic algorithm was used for parameter optimization. Through continuous attempts, the prediction effect was the best when the genetic algebra was 62, and the individual number was 45. The training network showed that the test set’s fitting degree was 96.07%, and the prediction error was 0.84%, which was much better than those before optimization. When the predicted risk value was more significant than 0.39, the bank should be vigilant and strengthen risk prevention. The development of internet finance can reduce commercial banks’ business risk levels, reduce their dependence on traditional business, and decrease commercial banks’ business risk levels. It can be seen that commercial banks can effectively improve risk management ability and efficiency promoted by technological development, so the level of business risk they undertake can be reduced.
Service provider portfolio selection (SPPS) can be a major challenge for organizations to achieve project success. Hence, organizations need to decide on which service provider portfolio (SPP) is appropriate for project management (PM).
However, there has been limited research on how to select a SPP in PM. To address this research gap, we establish a novel model for SPPS based on a BP neural network integrated with entropy-AHP from the perspective of the comprehensive economic benefit. This model employs a BP neural network due to its robustness and memory and nonlinear mapping abilities. Furthermore, we implement the proposed model for a construction project to verify the effectiveness. Our results indicate that the model performs well with a prediction accuracy of 97%. Moreover, the model is confirmed to be robust as it still achieves high prediction accuracy when the input data are disturbed randomly.
In order to improve the efficiency and accuracy of thermal and moisture comfort prediction of underwear, a new prediction model is designed by using principal component analysis method to reduce the dimension of related variables and eliminate the multi-collinearity relationship between variables, and then inputting the converted variables into genetic algorithm (GA) and BP neural network. In order to avoid the problems of slow convergence speed and easy falling into local minimum of Back Propagation (BP) neural network, this paper adopted GA to optimize the weights and thresholds of BP neural network, and utilized MATLAB software to program, and established the prediction models of BP neural network and GA–BP neural network. To verify the superiority of the model, the predicted result of GA–BP, PCA–BP and BP are compared with GA–BP neural network. The results show that PCA could improve the accuracy and adaptability of GA–BP neural network for thermal and moisture comfort prediction. PCA–GA–BP model is obviously superior to GA–BP, PCA–BP, BP, SVM and K-means prediction models, which could accurately predict thermal and moisture comfort of underwear. The model has better accuracy prediction and simpler structure.
The genetic algorithm-backpropagation neural network algorithm (GA-BP) makes full use of the advantages of genetic algorithm (GA) and BP neural network (BPNN). It has been widely used in practical problems, but it also has shortcomings such as slow convergence. Inspired by the generative adversarial network, a population competition mechanism (PCM) is proposed to improve the search ability of the GA, two populations are generated to compete, and the winning population can obtain the optimal individual from the failed population to ensure that the winning population has a sustainable advantage. The failed population will randomly generate new individuals and add new genes so as to get better individuals, through such a mechanism to ensure the rapid optimization of the entire population, avoid the risk of premature convergence, speed up the iterative speed and improve the stability of the GA. According to the characteristics of the fitness landscape, the learning rate of the BPNN is optimized, and it can change adaptively, which can effectively improve the network convergence speed and greatly reduce the time cost. We define the GA-BP that improves the learning rate based on fitness landscape as FL-GA-BP algorithm. On the basis of the FL-GA-BP algorithm, adding GA improved with PCM, we define this new algorithm as I-GA-BP algorithm, namely the I-GA-BP algorithm that combines PCM-improved GA and fitness landscape-improved BPNN. In this paper, we use two types of test functions with different characteristics and complexity to conduct experiments to verify the effectiveness of the I-GA-BP algorithm. By comparing the experimental data of the three algorithms GA-BP, FL-GA-BP and I-GA-BP, it is obtained that the I-GA-BP algorithm can better escape from the local optimal solution, which is more conducive to finding the global optimal solution. It also greatly improves the convergence speed of the neural network. Finally, we briefly discussed the effect of adjusting the number of neurons on the stability of the I-GA-BP algorithm.
The purpose of this article is to present a structured review of publications utilizing machine learning methods to aid in engineering risk assessment. A keyword search is performed to retrieve relevant articles from the databases of Scopus and Engineering Village. The search results are filtered according to seven selection criteria. The filtering process resulted in the retrieval of one hundred and twenty-four relevant research articles. Statistics based on different categories from the citation database is presented. By reviewing the articles, additional categories, such as the type of machine learning algorithm used, the type of input source used, the type of industry targeted, the type of implementation, and the intended risk assessment phase are also determined. The findings show that the automotive industry is leading the adoption of machine learning algorithms for risk assessment. Artificial neural networks are the most applied machine learning method to aid in engineering risk assessment. Additional findings from the review process are also presented in this article.
Project-based organizations have their processes based on project, program and portfolio approach, they work within a multi-project environment and their core activities are carried out through project execution under an integrated organizational project management. Therefore, operational risks in project-based organizations are directly related to the project risk management. Thus, in project-based organizations risk management is not only about project risk management, because literature also shows that risk management applied to each project in an isolated way is not enough for systemic risk management in project portfolio and organizational context. This paper, based on a systematic literature review on project portfolio risk management, identifies project portfolio risk categories proposed in previous studies, and proposes and describes four categories associated to operational risk in project portfolio context: project portfolio management level, environmental conditions, project interactions and internal processes. The categories proposed for operational risk in project-based organizations under project portfolio approach are oriented to risk management focused not on the operational risk of each isolated project, but on the project portfolio management. The results can be considered a partial contribution towards building conceptual elements to support project portfolio management process.
In many cases of practical multiattribute project portfolio selection problems, it is hard to obtain accurate measurements of attributes and precise preference information. Even after a long and costly information gathering, the attribute measurements and the preference information can still be uncertain or inaccurate. Considerable cost saving will be obtained if the selection of an optimal project portfolio can be done using rank‐level information based on some or all the attributes, without knowing the preference information. In this paper, we propose a stochastic multiattribute acceptability analysis‐based method that can deal with mixed rank and cardinal attribute measurements and uses little or no weight information. In the proposed stochastic multiattribute acceptability analysis‐based method, the decision makers need not to express their preferences explicitly or implicitly, so it is particularly useful when no weight information is available at all. A numerical example involving selection of photovoltaic plants in an industrial province in Eastern China is provided to demonstrate the proposed method.
Even though systems thinking has been highlighted in portfolio management theory, independent project control logic still dominates its implementation process. This paper constructs a system dynamics model for portfolio monitoring and control. Considering the on-going portfolio as a complex social system, the impacts of project interdependencies ( ) on portfolio decision-making are investigated under a behavioral paradigm. Our findings indicate the remedial actions, affected by behavioral factors like planning fallacy and ‘Pet project’ effects, may generate escalation of commitment under specific levels of uncertainty and interdependencies. Thus, portfolio coordination decisions should be made from a strategic perspective with the consideration of complexities embedded in the system and behavioral responses from portfolio managers.
Project management includes several managements, such as time management, cost management, and quality management…etc. Project risk management is one of the most important management, especially in this time, which has many unexpected events.
This management means with classification, analyzing, planning, identification, assessment, and response and avoidance strategies of risks.
Therefore, it should be too interested in risk management, to avoid many losses.
This paper focuses on project portfolio risk categorisation. Based on literature, a list including risks characteristic for a project portfolio was developed. After the assessment procedure (using the Delphi method), when the expert consensus had been achieved, thirty-six project portfolio risks were selected. The applied research procedure assumed project portfolio risk assessment, according to the approach suggested in the literature of the subject, including the likelihood of a given risk. During the research work, for the data clustering, exploratory factor analysis (EFA) has been applied. As a result of the analysis, a taxonomy of project portfolio risks was developed and a risk category with the greatest likelihood of occurrence was indicated.
Aiming at the problem that the most existing fault diagnosis methods could not effectively recognize the early faults in the rotating machinery, the empirical mode decomposition, fuzzy information entropy, improved particle swarm optimization algorithm and least squares support vector machines are introduced into the fault diagnosis to propose a novel intelligent diagnosis method, which is applied to diagnose the faults of the motor bearing in this paper. In the proposed method, the vibration signal is decomposed into a set of intrinsic mode functions (IMFs) by using empirical mode decomposition method. The fuzzy information entropy values of IMFs are calculated to reveal the intrinsic characteristics of the vibration signal and considered as feature vectors. Then the diversity mutation strategy, neighborhood mutation strategy, learning factor strategy and inertia weight strategy for basic particle swarm optimization (PSO) algorithm are used to propose an improved PSO algorithm. The improved PSO algorithm is used to optimize the parameters of least squares support vector machines (LS-SVM) in order to construct an optimal LS-SVM classifier, which is used to classify the fault. Finally, the proposed fault diagnosis method is fully evaluated by experiments and comparative studies for motor bearing. The experiment results indicate that the fuzzy information entropy can accurately and more completely extract the characteristics of the vibration signal. The improved PSO algorithm can effectively improve the classification accuracy of LS-SVM, and the proposed fault diagnosis method outperforms the other mentioned methods in this paper and published in the literature. It provides a new method for fault diagnosis of rotating machinery.
In traditional information technology project portfolio management (ITPPM), managers often pay more attention to the optimization of portfolio selection in the initial stage. In fact, during the portfolio implementation process, there are still issues to be optimized. Organizing cooperation will enhance the efficiency, although it brings more immediate risk due to the complex variety of links between projects. In order to balance the efficiency and risk, an optimization method is presented based on the complex network theory and entropy, which will assist portfolio managers in recognizing the structure of the portfolio and determine the cooperation range. Firstly, a complex network model for an IT project portfolio is constructed, in which the project is simulated as an artificial life agent. At the same time, the portfolio is viewed as a small scale of society. Following this, social network analysis is used to detect and divide communities in order to estimate the roles of projects between different portfolios. Based on these, the efficiency and the risk are measured using entropy and are balanced through searching for adequate hierarchy community divisions. Thus, the activities of cooperation in organizations, risk management, and so on—which are usually viewed as an important art—can be discussed and conducted based on quantity calculations.
To facilitate the project portfolio benefits (PPB) management and realize the maximization of the benefits, a PPB measurement model based on the genetic algorithm (GA)-BP neural network group (GA-BPNNG) is proposed in current research. Unlike traditional benefits management approaches, the proposed model takes full cooperation and information sharing advantages of group learning to measure PPB more accurately, which considers both the financial and nonfinancial benefits and synergy benefits generated from the interrelationships among project portfolio components. To ascertain the priority of the PPB measurement model, the GA-BPNNG model is compared with BPNN and GA-BPNN, two common models in the literature. The mean square error of the GA-BPNNG model reduced by 94% and 83%, respectively, when compared to the BPNN and GA-BPNN models. The results suggest that the PPB measurement model performs more effectively. Therefore, it could be concluded that the proposed model has a better nonlinear fitting effect for PPB measurement. This article can support managers, decision-makers, and management in realizing strategy by providing a practical and scalable tool for PPB measurement. Index Terms-Benefits management (BM), genetic algorithm (GA)-BP neural network group (GA-BPNNG), project portfolio (PP), project portfolio benefits (PPBs) measurement, synergy benefits.
Project portfolio networks (PPNs) are becoming increasingly interdependent at multiple levels to better realize enterprises’ strategic objectives. However, interdependencies between PPN components increase the potential for risk cascading propagation and amplify the impact of certain risks into the collapse of the entire PPN. To maintain PPN robustness and controllability, it is necessary to study the risk cascading propagation process in PPN. First, the PPN is constructed based on a scale-free network generation algorithm. Second, a risk cascading propagation model of the PPN is developed using a load capacity model. Four different mixed strategies are proposed to explore the effect of risk cascading propagation on the PPN, and a numerical simulation is applied to examine the robustness and controllability of the PPN under different strategies. The results show that the nonlinear load capacity model with the initial residual capacity redistribution principle provides optimal robustness against risk cascading propagation. When the management capacity redundancy reaches a certain value, PPN has better risk cascading propagation controllability, regardless of the mixed strategy. These results provide a basis for managers to protect against risk cascading propagation and improve PPN robustness and controllability in a specific context. The proposed model can also enrich project portfolios and risk management theory.
Project portfolio risks (PPRs) are mostly considered in terms of interdependency between projects, ignoring the time dependency and causality between risks. This may lead to inappropriate risk assessments and reduced efficacy in risk treatments. This study aims to dynamically assess PPRs from the life cycle perspective to support managers in planning risk treatment actions more effectively. A fuzzy dynamic Bayesian network (F-DBN) is applicable in this scenario. First, PPRs are identified by considering project interdependency. Second, the causality and time dependency between the PPRs are modeled using an F-DBN. Then, the dynamic variation characteristics of the PPRs in the project portfolio (PP) life cycle are revealed. Finally, a numerical example is adopted to validate the applicability and effectiveness of the model. Based on the results, the key PPRs at different stages are identified, and the inherent characteristics with dynamic changes of those key risks are further revealed. The results of the analysis provide insights for PP managers to implement corresponding risk-reduction strategies.
Transportation of live sheep is a relatively important part of the meat sheep industry, and the generation of transportation stress seriously threatens the organism health of live sheep as well as the quality of lamb meat. In this study, we selected Lake sheep, one of the edible sheep breeds, as the meat sheep research object, analyzed the actual needs of meat sheep transportation bio-sensing and modeling evaluation, and built a systematic bio-signal detection, processing and modeling method from bio-signal sensing, core waveform extraction to feature parameter estimation using wearable meat sheep physiological dynamic sensing and continuous monitoring of transportation environment sensors, and used modern spectrum estimation methods to judge and strip the target signal waveform from it to achieve accurate sensing and acquisition of transport key parameters. We also use modern spectral estimation methods to determine and extract the target signal waveforms from them, and achieve accurate sensing and acquisition of transport key parameters. Based on this, a qualitative stress assessment method based on external performance and a graded prediction model for transportation stress of meat sheep are constructed, and a predictive analysis and estimation management mechanism of stress level based on genetic algorithm optimized reverse artificial neural network is realized. Using a qualitative stress assessment method based on external manifestations, we established stage-specific classification rules for monitoring data in the transportation environment of meat sheep.The artificial neural network with optimized parameters based on the classification rules applying genetic algorithm is designed and implemented for real-time and efficient stress level prediction management mechanism for meat sheep transportation. The graded prediction model of meat sheep transportation stress was established by GA-BPNN and compared with the BP neural network optimized by particle swarm algorithm, simulated annealing algorithm and ant colony algorithm. The best optimization effect of GA was obtained, and the average prediction accuracy of the model could reach 89.81%. Finally, it achieves to improve the transportation reliability, reduce the transportation risk, and solve the problems of inefficient meat sheep transportation supervision and quality control.
Purpose
Project portfolio risk (PPR) management plays an important role in promoting the smooth implementation of a project portfolio (PP). Accurate PPR prediction helps managers cope with risks timely in complicated PP environments. However, studies on accurate PPR impact degree prediction, which consists of both risk occurrence probabilities and risk impact consequences considering project interactions, are limited. This study aims to model PPR prediction and expand PPR prediction tools.
Design/methodology/approach
In this study, the authors build a PPR prediction model based on a genetic algorithm and back-propagation neural network (GA-BPNN) integrated with entropy-trapezoidal fuzzy numbers. Then, the authors verify the proposed model with real data and obtain PPR impact degrees.
Findings
The test results indicate that the proposed method achieves an average absolute error of 0.002 and an average prediction accuracy rate of 97.8%. The former is reduced by 0.038, while the latter is improved by 32.1% when compared with the results of the original BPNN model. Finally, the authors conduct an index sensitivity analysis for identifying critical risks to effectively control them.
Originality/value
This study develops a hybrid PPR prediction model that integrates a GA-BPNN with entropy-trapezoidal fuzzy numbers. The authors use this model to predict PPR impact degrees, which consist of both risk occurrence probabilities and risk impact consequences considering project interactions. The results provide insights into PPR management.
In the present study, the air-overpressure (AOp) due to mine blasting is predicted using an uncertainty intelligence method based on the Z-number reliability and fuzzy cognitive map (FCM). Hence, the cause-and-effect relationship between effective parameters on AOp was first determined by a team of experts. But, expert opinions are accompanied by uncertainty. Therefore, Z-number concept is used to solve the uncertainty of expert opinions and improve views. Notably, the relationships dependent on expert opinions. To overcome this problem, two learning algorithms called nonlinear Hebbian (NLH) and differential evolution (DE) algorithms integrated to decrease the dependence on experts’ views. An FCM based on NLH-DE and reliability information was first designed; then inputs’ weights were extracted during several simulations in training process. The reliability inputs’ weights were imported into back-propagation causality-weighted neural networks (BPCWNNs). The coefficient of determination (R²), root mean square error (RMSE), variance account for (VAF), and accuracy indices were employed as evaluations criteria to select the optimal topology of artificial neural networks (ANNs). Back propagation neural network (BPNN) and generalized feed forward neural network (GFFNN) models were also used to compare and analyze the performance of the developed BPCWNNs model. The results indicated that Z-number considerably improved the accuracy of the ANN model. Notably, the model assessment revealed that the proposed BPCWNNs model was accurate compared to the BPNN and GFFNN and selected as the most superior model. The BPCWNN is the model with high performance which can be used as an artificial intelligence technique for predicting blast-induced AOp.
It is a trend for a production system to produce multiple product types with small batches. Thus, there are frequent switches between batches. Due to the different product types in different batches, setups may be necessary for a batch changeover. Based on practical applications, we investigate the problem of scheduling a class of two-stage hybrid flow shops, where there is significant setup time at the first stage, while the second stage should operate continuously without interruption. To solve this problem, after deriving the existence conditions of a feasible schedule, an adaptive genetic algorithm is designed to dynamically adjust the probabilities of crossover and mutation operations according to the population diversity. In this way, the whole solution space could be fully explored so as to get an optimal or near optimal solution for large-size practical problems. Also, based on the characteristics of solutions, a local search method is presented to further improve the solution accuracy. Experiments are done to test the efficiency and effectiveness of the proposed method. Results show that the algorithm can find good solutions with a gap of 1% with the lower bound within three minutes, i.e., it is efficient and effective.
Due to the lack of historical data, experts often score risks based on their experience and some linguistic terms in risk assessment, then the risk assessment is essentially a semi-quantitative problem. When experts score risks, the scores are possibly related to experts’ risk attitudes. On the other hand, the linguistic terms used are inevitably ambiguous, and interval-valued fuzzy numbers can deal with linguistic uncertainty better in complex situation. So a risk analysis model based on interval-valued fuzzy numbers and risk attitudes is novelly proposed in this paper. For safety risks in oil industry, the risk consequence often performs in several aspects, some of them are difficult to be measured by money and cannot be aggregated to a comprehensive index directly. A multi-expert and multi-criterion information fusion(MEMC-IF) model is needed. Firstly, linguistic terms and interval-valued fuzzy numbers are determined and a MEMC-IF model is constructed to derive the collective data and the comprehensive risk consequence. Secondly, a defuzzification model is presented to transform interval-valued fuzzy numbers to crisp values with considering risk attitudes novelly. Then, a risk matrix is constructed to assess which risks are serious and which risks can be ignored. In addition, a case study is demonstrated to show the efficiency of the proposed model and a discussion is completed.
There are numerous projects with many interdependencies in the new product development (PD) portfolio, which can be viewed as a complex network. Due to the dependency among projects, the risks in one project may transmit to others, which will ultimately affect the success of the entire project portfolio. Thus, evaluating the priority and criticality of projects under the influence of risk propagation is a fundamental challenge for the PD project portfolio. This paper presents a risk propagation model using the Bayesian network and brittle risks. The criticality analysis presented is based on both the local and the global effects of risk in the portfolio. Firstly, we build a Bayesian network model of the portfolio based on the analysis of relationships between projects. Further, we measure the criticality of the project based on the local impact of risk propagation using Bayesian and complex networks. Then, we measure the criticality of the project based on the global impact of risk propagation using brittleness. Finally, we synthesize the calculation results of criticality to measure the integrated criticality of the portfolio, and examine an exemplar industrial portfolio to verify the effectiveness of the proposed model and method.
In mining supply chains, transporting ore from mines to ports, is a problem of significant interest. Resources required in transporting ore are limited, and hence using resource efficiently goes a long way towards solving this problem. A key issue is that the resource availability is subject to uncertainties. Previous studies have considered similar (deterministic) problems, commonly tackled as resource constrained job scheduling. However, the uncertainty associated with resources needed to transport ore has not been considered. This study extends previous ones, by investigating resource constrained job scheduling with uncertainty (considering resource availabilities). The focus is on robust optimisation, where the aim is to find high-quality solutions irrespective of the input data. To solve this problem a population-based ant colony optimisation approach is devised, with the aim of identifying high-quality solutions across several uncertain scenarios. Moreover, computing the objective is inefficient due to uncertainty, and hence surrogate models are used. Experiments are conducted on a wide range of problem instances with varying uncertainty levels considering resources, and find that population-based ant colony optimisation is superior to ant colony optimisation on its own. The key advantage of this method is that it is able to maintain a population of excellent solutions, which can be used to efficiently approximate the objective values of new solutions, and hence update the pheromones globally of the ant colony optimisation algorithm. Using this information the surrogate model considers only high quality solutions, and by carrying out increased learning in short time-frames, achieves excellent results.
Purpose – Comprehensive project portfolio risk (PPR) analysis is essential for the success and sustainable development of project portfolios (PPs). However, project interdependency creates complexity for PPR analysis. In this study, considering the interdependency effect among projects, the authors develop a quantitative evaluation model to analyze PPR based on a fuzzy Bayesian network.
Design/methodology/approach –In this paper, the primary purpose is to comprehensively evaluate project portfolio risk considering the interdependency effect using a systematical model. Accordingly, a fuzzy Bayesian network (FBN) is developed based on the existing studies. Specifically, first, the risks in project portfolios are identified from the project interdependencies perspective. Second, a fuzzy Bayesian network is adopted to model and quantify the interaction relationships among risks. Finally, the model is implemented to analyze the occurrence situation and characteristics of risks.
Findings –The interdependency effect can lead to high-stake risks, including weak financial liquidity, a lack of cross-project members and project priority imbalance. Furthermore, project schedule risks and inconsistency between product supply and market demand are relatively sensitive and should also be prioritized. Also, the validity of this risk evaluation model has been proved.
Originality/value –The findings identify the most sensitive risks for guaranteeing portfolio implementation and reveal interdependency effect can trigger some specific risks more often. This study proposes for the first time to measure and analyze project portfolio risk by a systematical model. It can help systematically assess and manage the complicated and interdependent risks associated with project portfolios.
Strategic risk is an inevitable question of reality, which leads to a significant impact that negatively affects firms’ overall development, even threatening their survival. Most of extant studies on early warning models merely take the financial factors into account, whereas strategic risk pertaining to firms involves a diverse group of non-financial factors. In addition, the growing attention attached to the sustainable development of ecological and social environment has raised specific concerns to manufacturing firms, which need especially manage the risks generated from the process of sustainable innovation. However, scholars have paid far less attention to early warning models that comprehensively address the above problems. It is crucial to facilitate manufacturing firms to timely detect strategic risk during the process of sustainable innovation, whereas the existing early warning models are not reliable enough to warn the potential risks. Therefore, this study sets out to introduce the sustainable risk into the early warning indicators system of strategic risk. Then, Chinese manufacturing firms are applied to construct an early warning model which based on the back-propagation (BP) neural network optimized by genetic algorithm (GA). The results provide evidence that the proposed model shows a better fitting effect, higher prediction precision and higher convergence speed when compared with conventional models. In addition, the weights of the early warning model obtained from the training are substituted into the forward propagation formula, which enables to determine the relative importance of risk factors. It is of great significance for the firms to prioritize risk management actions. Furthermore, the new model enables manufacturing firms to warn the strategic risk during the process of sustainable innovation while striking a balance among economic, environmental, and social performances.
Micro milling is widely used to manufacture complex miniature structure with high quality, and the related sustainability evaluation is a complicated multi-factor decision-making problem. This paper proposes an intelligent system for evaluating sustainability performance of micro milling process based on the principal component analysis (PCA) algorithm and the back propagation (BP) neural network. As a critical influence factor of sustainability evaluation, the non-linear micro cutting tool life can be predicted by integrating the particle filter (PF) algorithm and the long short-term memory (LSTM) network based on the stochastic tool wear. Then, the systematical sustainability assessment metrics of micro milling process are analyzed in the environmental, economic and social perspectives. Considering the nonlinearity and complexity of sustainability evaluation, the intelligent integrated PCA-BP evaluation method is used to improve the calculation efficiency and simply the evaluation process, in which the dimension of multiple sustainability evaluation factors is reduced by the PCA algorithm. The micro milling experiments with workpiece material Al6061 were conducted to validate the feasibility of the proposed intelligent evaluation methodology. The intelligent sustainability evaluation results agree with the traditional weighted sustainability performance index analysis on the basis of the manner “higher is better”. For the proposed intelligent integrated PCA-BP evaluation method, the training steps reduced from 65 times to 38 times and the prediction accuracy increased from 82.57% to 90.59% compared to the traditional BP network. The comparison results showed that the proposed intelligent integrated PCA-BP evaluation method can obtain the sustainability evaluation value automatically with high efficiency and practicability, and it also provides the decision-making base for the micro milling process optimization.
Mining industry is the basic industry of the national economy. However, in recent years, listed mining companies have suffered serious financial risks due to special reasons such as poor spot market liquidity of their products, strong policy dependence, and long investment payback periods. In the previous studies, most of the financial crisis prediction focused on the whole industry and manufacturing industry. The research on the financial risk of mining enterprises focuses more on how to adjust R&D activities, environmental performance to improve the financial performance of enterprises. There is still a lot of room for in-depth research on the systematic prevention and early warning of financial risks of listed mining companies. At the same time, in terms of research methods, many scholars used multivariate discriminant model, logistic regression model and support vector machine model. Compared with the Back-Propagation (BP) neural network model, these model methods have more or less defects. Therefore, we take mining listed companies as the research object, select the financial data of China's A-share mining listed companies in 2018, and construct the BP neural network financial early warning model, trying to provide more practical means for the financial risk early warning of mining companies. The research conclusions of this paper are as follows: (1) The BP neural network financial early warning model constructed in this paper has high prediction accuracy, which can be well used in the practice of financial early warning of mining listed companies; (2) The financial situation of China's A-share mining listed companies in 2018 is generally in a good state. The companies with good financial status can effectively control the cost and have good debt paying ability while earning income; (3) For companies with financial status that require early warning, the root cause is mainly that they do not pay attention to the risk of bad debt losses, which makes current credit sales income and accounts receivable are at high levels, and they also do not have good profitability.
An increase in the penetration of renewable energy sources in the electrical production has been matched by the emergence of many and varied challenges and problems. Among the most important challenges is finding the smart technologies and algorithms which are capable of achieving efficient solutions. This chapter provides an expanded view of the uses of the particle swarm optimization (PSO) algorithm in the renewable energy systems field. Additionally, it describes how the algorithm can be developed to cope with problems related to renewable energies to achieve desired goals. The PSO algorithm was used to solve many problems in the renewable energy systems, such as in optimal hybrid power systems, optimal sizing, and optimal net present cost, among others, where the PSO algorithm showed its high adaptability in problem-solving. Further, many researchers proceeded with the study and development of the PSO algorithm. In contrast, other researchers tried to hybridize it with different algorithms to be more efficient and convenient to overcome some of the problems and challenges that they encountered. The renewable energy systems have several issues to discuss, such as the cost of investment, the feasible technical criteria, optimal control, and the ecological problems as well as the social effect. Overall, studies and research have proven that the PSO algorithm is one of the best algorithms used in the field of renewable energy. This is attributed to the algorithm’s simplicity, high efficiency, and effectiveness compared to other algorithms and optimization methods.
We consider a location-inventory optimization model for supply chain (SC) configuration. It includes a supplier, multiple distribution centers (DCs), and multiple retailers. Customer demand and replenishment lead time are considered to be stochastic. Two classes of customer orders, priority and ordinary, are assumed based on their demand. The goal is to find the optimal locations for DCs and their inventory policy simultaneously. For this purpose, a two-phase approach based on queuing theory and stochastic optimization was developed. In the first phase, the stock level of DCs is modeled as a Markov chain process and is analyzed, while in the second phase, a mathematical program is used to determine the optimal number and locations of DCs, the assignment of retailers to DCs, and the order quantity and safety stock level at DCs. As solving this problem is NP-hard, a hybrid Genetic Algorithm (GA) was developed to make the problem computationally tractable.
In order to improve the risk evaluation and management in fresh grape supply chain and enhance the sustainable level of the supply chain, this study applied neural network to evaluate the risk of fresh grape supply chain from the perspective of sustainable development. Firstly, the possible risk factors in the supply chain were identified and the risk evaluation index system were proposed; then risk evaluation models based on single BP and optimized BP (GABP and PSO-BP) neural network were established; and then the models were trained, tested and evaluated using data set from supply chain survey. The survey and analysis results showed that the risk of fresh grape supply chain was at a low level but the risk in each link was discrepant, the biggest risk was the risks among the links in the chain (R0), and the high risk dimensions were the economic risk, social risk and cooperation risk; most risky events were located in the second quadrant (small probability & high damage risk events). The results of models training and testing indicated that the optimized model was superior to single BP neural network for risk assessment in grape sustainable supply chain, and the PSO-BP model was more accurate and suitable with less evaluation errors and a bigger R2. The results also extracted the risk factors that contributed most to the overall risk of grape sustainable supply chain. This paper enriches the method of supply chain risk assessment theoretically, and provides practical suggestions for risk prevention, stable operation and sustainability improvement of fresh grape supply chain.
Feature Extraction Algorithms (FEAs) aim to address the curse of dimensionality that makes machine learning algorithms incompetent. Our study conceptually and empirically explores the most representative FEAs. First, we review the theoretical background of many FEAs from different categories (linear vs. nonlinear, supervised vs. unsupervised, random projection-based vs. manifold-based), present their algorithms, and conduct a conceptual comparison of these methods. Secondly, for three challenging binary and multi-class datasets, we determine the optimal sets of new features and assess the quality of the various transformed feature spaces in terms of statistical significance and power analysis, and the FEA efficacy in terms of classification accuracy and speed.
Link of the paper: https://authors.elsevier.com/a/1cc%7E%7E5buwWjZZw
This paper presents a brief review on major accidents and conducts bibliometric analysis of risk assessment methods for excavation system in recent year. The summarization of potential risks during excavation provides an important index for establishing an early warning system. The applications of fuzzy set theory and machine learning methods in risk assessment during excavation are presented. A case study of excavation in Guangzhou metro station is used to demonstrate the application of a machine learning method for risk evaluation. The large amount of data collected by 3S techniques (RS, GIS and GPS) and sensors increases accuracy of risk assessment levels in excavation. These procedures, integrated into building information modelling (BIM) management platform, can manipulate dynamic safety risk monitoring, control, and management. Finally, the processing and analysis of big data obtained from 3S techniques and sensors provide promising perspectives for establishing integrated technology system for excavation.
In the era of big data, it is aimed to use big data technology to form an effective early warning and prevention of Internet credit. The BP neural network algorithm is applied to determine the number of nodes, activation function, learning rate, and other parameters of each layer of the BP neural network. Also, many data samples are used to build an early warning model of Internet credit risk. The constructed model is trained and tested. Finally, the genetic algorithm (GA) is used to optimize the neural network to improve the accuracy of early warning. The results show that based on 450 data samples from 90 enterprises over five years and the risk interval divided by the “3σ” rule, the Internet credit risk level is initially determined. Then, the neural network is trained and tested. The network output rate is as high as 85%. To avoid the defect of the BP neural network falling into local extreme value, GA is used to optimize the neural network. The warning is more accurate, and the error is smaller. The accuracy rate can reach 97%. Therefore, the use of BP neural network for early warning and assessment of Internet credit risk has good accuracy and computing efficiency, which expands the application of BP neural network in the field of Internet finance, and provides a new development direction for the early warning and assessment of Internet credit risk.
Peer-to-peer (P2P) lending platform plays a significant role in modern financial systems. However, due to improper supervision, credit risk is inevitable. In this paper, we analyze the traditional financial risk and information technology risk of P2P lending platform. In order to evaluate the performance of assessment algorithms, we present a BP neural network-based algorithm for lending risk assessment. To achieve our task, we crawled large-scale lending data for 2015-2019. Logistic regression is used to compare with BP neural network method. Experimental results show that BP neural network-based algorithm outperforms traditional Logistic regression algorithm and the proposed method can effectively reduce investor risk.
Unsupervised classification is a highly important task of machine learning methods. Although achieving great success in supervised classification, support vector machine (SVM) is much less utilized to classify unlabeled data points, which also induces many drawbacks including sensitive to nonlinear kernels and random initializations, high computational cost, unsuitable for imbalanced datasets. In this paper, to utilize the advantages of SVM and overcome the drawbacks of SVM-based clustering methods, we propose a completely new two-stage unsupervised classification method with no initialization: a new unsupervised kernel-free quadratic surface SVM (QSSVM) model is proposed to avoid selecting kernels and related kernel parameters, then a golden-section algorithm is designed to generate the appropriate classifier for balanced and imbalanced data. By studying certain properties of proposed model, a convergent decomposition algorithm is developed to implement this non-covex QSSVM model effectively and efficiently (in terms of computational cost). Numerical tests on artificial and public benchmark data indicate that the proposed unsupervised QSSVM method outperforms well-known clustering methods (including SVM-based and other state-of-the-art methods), particularly in terms of classification accuracy. Moreover, we extend and apply the proposed method to credit risk assessment by incorporating the T-test based feature weights. The promising numerical results on benchmark personal credit data and real-world corporate credit data strongly demonstrate the effectiveness, efficiency and interpretability of proposed method, as well as indicate its significant potential in certain real-world applications.
Based on the morphological information obtained from 2D slice images of real porous media, an improved simulated annealing algorithm (SAA) was proposed to reconstruct 3D large-scale porous media, which are intractable to handle for conventional SAA. Three different statistical functions were introduced to characterize the morphological information of real sandstone, including the one-point probability function, the two-point probability function and the lineal-path function. By changing the update method of the two-point probability function and the lineal-path function, i.e., using incremental calculation instead of conventional global calculation, the efficiency of reconstructing 3D large-scale porous media was greatly improved. Besides, in the later stage of reconstruction that the basic structure of porous media had been formed, the pixel selection algorithm was performed to speed up the reconstruction process. To evaluate the accuracy of the improved SAA, the similarity between the 3D reconstructed volume and the reference image of prototype sandstone was examined. The results showed good agreement between the reconstructed model and the references. The efficiency of the improved SAA was verified by comparison with the conventional SAA, the results of which indicated that the improved SAA can significantly shorten the reconstruction time of 3D large-scale porous media.
Cryptocurrencies, with Bitcoin being the most notable example, have attracted considerable attention in recent years, and they have experienced large fluctuations in their price. While a few studies employ conventional statistical and econometric approaches to reveal the driving factors of Bitcoin's prices, research on the development of forecasting models to be used as decision support tools in investment strategies is scarce. This study proposes a computational intelligence technique that uses a hybrid Neuro-Fuzzy controller, namely PATSOS, to forecast the direction in the change of the daily price of Bitcoin. The proposed methodology outperforms two other computational intelligence models, the first being developed with a simpler neuro-fuzzy approach, and the second being developed with artificial neural networks. Furthermore, the investment returns achieved by a trading simulation, based on the signals of the proposed model, are 71.21% higher than the ones achieved through a naive buy-and-hold strategy. The performance of the PATSOS system is robust to the use of other cryptocurrencies.
Construction projects face substantial hazards and losses due to their increasing complexity. Therefore, risk management plays an undeniable role in avoidance and mitigation of extended time and cost overruns. In this paper, a project portfolio risk assessment model based on the Bayesian network approach has been proposed in which the probabilities and expected values of schedule delays and cost overruns in construction projects are evaluated with respect to different risk levels of the project portfolio. The developed approach not only explicitly quantifies uncertainty in the cost and time of projects but also provides an appropriate method for modeling complex relationships in projects, such as causal relationships between risks as well as use of expert judgments for evaluating prior and conditional probabilities. The developed model is applied to Phases 13 and 14 of the South Pars gas field development projects in Iran as a case study to validate the model. Findings of the study reveal that shortage of resources, unforeseen activities, and contractor financial problems can definitely bring about delay in these projects.
Project Portfolio Management (PPM) is a modern management model and an instrument for evaluating, prioritizing and reviewing projects. To select a project portfolio that is a combination of these criteria, identifying the opportunities, evaluating the alignment between the project and the purposes and structure of the organization and analyzing the costs, benefits and risks of the project are essential. The present article uses a qualitative and quantitative approach for the optimal selection of project portfolios and applies a set of techniques, including data envelopment analysis (DEA) and fuzzy analytic network process (FANP) for project portfolio management space for the first time. This algorithm can also be used to optimize project portfolios based on a multi-objective non-linear programming (MONLP) model. The MONLP model will be solved by efficient branch and bound algorithm. A more accurate estimation can be accomplished for the selection of a low-risk portfolio by adding these two new criteria for computing risk numbers in terms of their proximity to the event and the ability to control and modify them in the future. This algorithm will ultimately be applied for a real case study in Mobarakeh Steel Co., a leading Iranian company producing steel sheets. © 2018, Bucharest University of Economic Studies. All rights reserved.
Strategies in project assortment and prioritization directly influence organizational production and cost-effectiveness. Due to dearth of funding and nominal technology with improper judgments of expert’s, selection of optimal project portfolio can be viewed as a risk based decision making problem considering various risk factors. Recently, fuzzy set theory is applied for modeling real world problems due to failure of traditional models in uncertain environments. This paper deals with project proposal prioritization from a set of project portfolio satisfying a set of criteria evaluated by decision makers. The analytic hierarchy process is applied in fuzzy environment to help the management of any project based company to prioritize in terms of investment. In the decision approach, expert team decides whether to accept or reject a project as per set of criteria and sub-criteria based on diverse project risk levels. A hierarchy among the different projects based on ratings clearly prioritize the projects among the suggested proposals.
In project portfolio selection, the aim is to choose projects which are expected to offer most value and satisfy relevant risk and other constraints. In this paper, we show that uncertainties about how much value the projects will offer, combined with the fact that only a subset of the proposed projects will be selected, lead to inaccurate risk estimates about the aggregate value provided by the selected project portfolio. In particular, when downside risks are measured in terms of lower percentiles of the distribution of portfolio value, these risk estimates will exhibit a systematic bias. For deriving unbiased risk estimates, we present a calibration framework in which the required calibration can be presented in closed-form in some cases or, more generally, derived by using Monte Carlo simulation to study a large number of project selection decisions. We also show that when the decision must comply with risk constraints, the introduction of tighter (more demanding) risk constraints can counterintuitively aggravate the underestimation of risks. Finally, we present how the calibrated risk estimates can be employed to align the portfolio with the decision maker's risk preferences while eliminating systematic biases in risk estimates. This article is protected by copyright. All rights reserved.