•527 articles related to imbalanced data and rare events are reviewed.•Viewing reviewed papers from both technical and practical perspectives.•Summarizing existing methods and corresponding ...statistics by a new taxonomy idea.•Categorizing 162 application papers into 13 domains and giving introduction.•Some opening questions are discussed at the end of this manuscript.
Rare events, especially those that could potentially negatively impact society, often require humans’ decision-making responses. Detecting rare events can be viewed as a prediction task in data mining and machine learning communities. As these events are rarely observed in daily life, the prediction task suffers from a lack of balanced data. In this paper, we provide an in depth review of rare event detection from an imbalanced learning perspective. Five hundred and seventeen related papers that have been published in the past decade were collected for the study. The initial statistics suggested that rare events detection and imbalanced learning are concerned across a wide range of research areas from management science to engineering. We reviewed all collected papers from both a technical and a practical point of view. Modeling methods discussed include techniques such as data preprocessing, classification algorithms and model evaluation. For applications, we first provide a comprehensive taxonomy of the existing application domains of imbalanced learning, and then we detail the applications for each category. Finally, some suggestions from the reviewed papers are incorporated with our experiences and judgments to offer further research directions for the imbalanced learning and rare event detection fields.
•Supply chain operates under competition at both the supply and demand markets.•Develop 3-stage game to study multi-sourcing and vertical information sharing.•Consider supply uncertainty, ...correlation, suppliers number and forecast error.•Examine tradeoff between risk diversification effect and price competition effect.•Investigate the interactions between multi-sourcing and vertical information sharing.
We study multi-sourcing and vertical information sharing (VIS) in a supply chain (SC), consisting of multiple suppliers with correlated supply uncertainty and two competing retailers with demand signal. The upstream and downstream SC members form a Stackelberg game, with the retailers being the leaders. The retailers engage in Cournot competition, while suppliers engage in Bertrand competition. We first study the incentive of VIS for retailers and find that retailers would not share information with suppliers for free. However, under side-payment contract, two possible equilibriums exist depending on whether it is complete VIS. Complete VIS approximates Pareto-optimality when retailers have accurate forecasts. If supplier correlation is high and the number of suppliers is large, suppliers would opt for low side-payment and expect higher forecast accuracy. When facing high supply uncertainty, the suppliers would seek lower side-payment and allow higher forecast error. Next, we examine the tradeoff between risk diversification effect (RDE) and price competition effect (PCE) facing retailers under different VIS strategies. We find if the number of suppliers is large and supplier correlation is low, then the retailers’ gain on RDE will exceed the loss on PCE. Same observation can be found when supplier number is low, retailers do not VIS, and supply uncertainty and forecast error meet certain conditions. Under VIS, retailers would choose highly-correlated suppliers to outsource. Subsequently, we derive the optimal strategy for the joint multi-sourcing and VIS scenario. Finally, numerical study is conducted to validate our models and derive managerial insights to enhance SC performance.
We study manufacturer-retailer bilateral information sharing in two competing supply chains (SCs), in which both the manufacturer and the retailer have partial information on demand. Based on ...Bertrand competition model and Winkler's consensus model, we develop a finite Bayesian Stackelberg game to analyze the two-way information sharing problem under horizontal supply chain (SC) competition. In line with the literature, we find that sharing demand forecast voluntarily in a SC benefits the manufacturer but hurts the retailer. However, we find whether SCs benefit from information sharing depends on competition intensity and forecast error. As competition is intensive, the expected values of information sharing (EVISs) for the entire SCs are high. Moreover, information sharing in one supply chain can improve the rival supply chain's EVIS under some conditions. Numerical experiments are conducted to get some managerial insights.
•Propose a two-way information sharing model under a supply-chain-level competition.•Investigate the equilibrium behavior of players in two competing supply chains.•Find competition can increase the value of information sharing for supply chain.•Explain how the information sharing in a supply chain impacts the rival's value.
► Develop AHP structure, capture relationship between various altruistic activities. ► Structure diverse elements in a hierarchy, and prioritize criteria and sub-criteria. ► Through comparisons, we ...arrange alternatives into homogeneous clusters. ► Synthesize priorities using pivot from one homogeneous cluster to an adjacent one. ► Relate widespread orders of magnitude of criteria and alternatives.
An innovative Analytic Hierarchy Process-based structure is developed to capture the relationship between various levels of activities contributed by people to society. Physical objects have widespread extension and degrees of importance that often differ by many orders of magnitude. Similarly, mental thoughts and criteria occur in widely heterogeneous entities that have to be sorted and arranged into homogeneous groups of few elements in each group so that one can evaluate the relationships among them accurately, from the smallest to the largest. It is through such a framework for organizing factors with smooth transition that it is possible to derive reliable priorities from expert judgments. The proposed model enables one to make decisions and allocate resources in as detailed and fine a way as possible. In addition to the traditional approach of structuring criteria into multiple clusters, the alternatives of a decision are also organized into the lowest multiple levels of that hierarchy. This arrangement and evaluation of alternatives differs from one criterion to another, which adds to the complexity of the undertaking when the alternatives are heterogeneous. The coherent approach to structuring complex decisions with the Analytic Hierarchy Process enables one to transcend the complexity of dealing in a scientific way with the problem of widespread orders of magnitude of criteria and alternatives in a complex decision. When the magnitudes are actually very small or very large, the accuracy of
rating
alternatives
one
at
a
time instead of comparing them in pairs involves much guessing, and can lead to a questionable outcome. Alternatively, comparisons, which are necessary for the measurement of intangibles, have greater and better justified accuracy.
Facing a customer market with rising demands for cloud service dependability and security, trustworthiness evaluation techniques are becoming essential to cloud service selection. But these methods ...are out of the reach to most customers as they require considerable expertise. Additionally, since the cloud service evaluation is often a costly and time-consuming process, it is not practical to measure trustworthy attributes of all candidates for each customer. Many existing models cannot easily deal with cloud services which have very few historical records. In this paper, we propose a novel service selection approach in which the missing value prediction and the multi-attribute trustworthiness evaluation are commonly taken into account. By simply collecting limited historical records, the current approach is able to support the personalized trustworthy service selection. The experimental results also show that our approach performs much better than other competing ones with respect to the customer preference and expectation in trustworthiness assessment.
•Apply Topic modeling to group synonyms under a topic to avoid human intervention and improve automatic market structure generation.•Develop the WVAP method to filter noises in Topic modeling results ...to elicit market structure.•Besides perceptual maps of product positioning, the proposed framework can provide rankings of products.
Studies have shown that perceptual maps derived from online consumer-generated data are effective for depicting market structure such as demonstrating positioning of competitive brands. However, most text mining algorithms would require manual reading to merge extracted product features with synonyms. In response, Topic modeling is introduced to group synonyms together under a topic automatically, leading to convenient and accurate evaluation of brands based on consumers’ online reviews. To ensure the feasibility of employing Topic modeling in assessing competitive brands, we developed a unique and novel framework named WVAP (Weights from Valid Posterior Probability) based on Scree plot technique. WVAP can filter the noises in posterior distribution obtained from Topic modeling, and improve accuracy in brand evaluation. A case study exploring online reviews of mobile phones is conducted. We extract topics to reflect the features of the cell phones with a qualified validity. In addition to perceptual maps derived by multi-dimensional scaling (MDS) for product positioning, we also rank these products by TOPSIS (Technique for Order Performance by Similarity to Ideal Solution) so as to visualize the market structure from different perspectives. Our case study of cell phones shows that the proposed framework is effective in mining online reviews and providing insights into the competitive landscape.
► Multi-Echelon supply chain with price and demand uncertainty. ► Integrated automobile manufacturing supply chain network. ► Trade-offs between inventories, production costs and customer service ...level. ► Fuzzy sets; stochastic chance-constrained programming; evolutionary computations.
This research is motivated by an automobile manufacturing supply chain network. It involves a multi-echelon production system with material supply, component fabrication, manufacturing, and final product distribution activities. We address the production planning issue by considering bill of materials and the trade-offs between inventories, production costs and customer service level. Due to its complexity, an integrated solution framework which combines scatter evolutionary algorithm, fuzzy programming and stochastic chance-constrained programming are combined to jointly take up the issue. We conduct a computational study to evaluate the model. Numerical results using the proposed algorithm confirm the advantage of the integrated planning approach. Compared with other solution methodologies, the supply chain profits from the proposed approach consistently outperform, in some cases up to 13% better. The impacts of uncertainty in demand, material price, and other parameters on the performance of the supply chain are studied through sensitivity analysis. We found the proposed model is effective in developing robust production plans under various market conditions.
The rapid development of e-commerce has expedited knowledge growth in the e-commerce social community. Knowledge sharing among online users has exhibited a nonlinear dynamic evolution. This paper ...examines the evolutionary process of knowledge sharing among users of the social commerce; builds an evolutionary game model to depict knowledge sharing phenomenon in the virtual community; and develops a mixed learning algorithm based on individual user’s historical game strategy, neighborhood user’s strategy, and information noise. We design a computational model based on multi-agent theory and social network, and implement computational experimental system using NetLogo 5.0. We find that the proposed computational–experimental model can help decision makers simulate evolutionary process under various scenarios. The evolutionary game rule and social network structure significantly influence the degree of cooperation and knowledge sharing among users. The greater noise the network information has the less stable the users’ behavior will be. One can thus identify an optimal initial cooperation rate to facilitate the system to reach equilibrium state quickly. Our study on the dynamic evolution of knowledge sharing behavior in the social commerce contributes to the theoretical development of literature and provides valuable decision-making support to managers.
We study a dual-channel recycling closed-loop supply chain (CLSC) and investigate
the royalty strategy involving cost-reducing technique for remanufacturing patented products. Facing
information ...asymmetry and market uncertainty, we address the problem where the patent licensor (manufacturer)
and licensee (remanufacturer) simultaneously compete in the sales market and the recycling market.
We examine the optimal decisions of a decentralized CLSC (D-CLSC) with the manufacturer being the
Stackelberg leader. Numerical examples are used to demonstrate how the patented technology (cost-reducing
technique) affects the channel players’ behaviors and how to identify the optimal royalty fee. Based on
the theoretical derivation and the numerical outcomes, we find that regardless of the CLSC structure
(centralized or decentralized), the take-back prices and the total profits will rise with the increase
of savings from the licensed technology. In the D-CLSC, (i) the expected profits of the manufacturer
and the remanufacturer as well as the royalty fee will also rise with the savings from the licensed
technology. (ii) In addition, the wholesale price, retail price, take-back prices, as well as the
royalty fee will rise with the degree of information asymmetry. But the retailer’s expected profit
will decline. (iii) Finally, the expected profit of the manufacturer will rise with the demand uncertainty
and the return uncertainty. For the remanufacturer, this trend is not obvious. Our research provides
guidance to resolve conflicts and intellectual property disputes between the original manufacturer
and the remanufacturer of the patented product.
This research aims at finding the best governing policy for offshore outsourcing of business activities. We use Analytical Network Process, a multicriteria decision making methodology, to create the ...evaluation framework. From the perspective of decision makers, stakeholders, and influence groups, four policy options are evaluated with respect to approximately 50 economic, political, technological and other factors. The model provides both long-term and short-term views of the outsourcing issue concerned to all parties. The all-inclusive approach helps policy makers to decide on the best policy and has the potential to ease tension between proponents and opponents of offshore outsourcing.