Warranty claims forecasting plays an increasingly important role not only for preparing financial plans but also for optimizing warranty policy and improving after-sale services. In the case of new products, an important feature is that the new generation of products often has a close connection with the previous generations of products it replaces. Thus, the warranty claims data of the previous generations of products can be used for extracting reliability information of new products. In this context, we propose a warranty claims forecasting model considering usage rate for new products sold with a two-dimensional warranty. The accelerate failure time model is introduced to investigate the effect of usage rate on product degradation. The non-homogeneous Poisson process is used to model failure counts of repairable products and the constrained maximum likelihood estimation method is used to estimate model parameters. The results of data experiments based on both simulation and real data collected from an automobile manufacturer in China show that the proposed model considering the varying usage rate outperforms the traditional models in forecasting the number of warranty claims.

As a high-energy-consumption and high-CO2-emission industry in China, the transportation sector has been under increasing pressure to improve its performance. This paper develops a novel parallel DEA approach to measure Chinese transportation sector’s energy and environmental performance (EEP) over all possible weights, which is to avoid the risk of using the extreme or the most favorable weights in performance evaluation. In our method, the transportation sector is consisted of two parallel subsystems (passenger transportation and freight transportation) with shared inputs and undesirable shared outputs. All possible weights are considered in the EEP evaluation, then the EEP of a transportation sector is represented by a ranking interval. Finally, the proposed approach is applied to the transportation sectors in 30 Chinese provinces. Results show that the best and the worst ranking of most provinces vary greatly. Besides, the EEP of most provinces is hard to dominate others strictly, but the general tendency is the EEP of eastern provinces better than western provinces. Furthermore, the EEP difference of some adjacent provinces with similar features is distinct. These findings are not all the same as previous studies, which verifies the necessity of considering all possible weights in performance evaluation. Therefore, our method provides a new point of view in performance evaluation and can give more robust results for decision makers.

The uncertainty distribution can more effectively express the uncertainty of decision makers’ judgments during a pairwise comparison of any alternatives. This paper investigates the priority models of group intuitionistic fuzzy preference relations with normal uncertainty distribution. The mathematical equivalence between the membership, non-membership degree interval fuzzy preference relation and the intuitionistic fuzzy preference relation is constructed, showing that there exists an inverse relationship between the priority of alternatives using these two types of interval preference relations. The new optimal models regarded the event that the deviation between the ideal judgement meeting the multiplicative consistency and the actual judgement obeying normal uncertainty distribution shall not exceed a threshold value under the given belief degree as a constraint, and regarded the minimum sum of all the threshold values as the objective function. The chance constraint was introduced to measure the degree to which multiplicative consistency can be realized under different belief degrees. The priority model provides a new method for simulating uncertainty and fuzziness in the real-world decision making environment.

In this paper, we consider an MMPP/M/1/1 retrial queue where incoming fresh calls arrive at the server according to a Markov modulated Poisson process (MMPP). Upon arrival, an incoming call either occupies the server if it is idle or joins a virtual waiting room called orbit if the server is busy. From the orbit, incoming calls retry to occupy the server in an exponentially distributed time and behave the same as a fresh incoming call. After an exponentially distributed idle time, the server makes an outgoing call whose duration is also exponentially distributed but with a different parameter from that of incoming calls. Our contribution is to derive the first order (law of large numbers) and the second order (central limit theorem) asymptotics for the distribution of the number of calls in the orbit under the condition that the retrial rate is extremely low. The asymptotic results are used to obtain the Gaussian approximation for the distribution of the number of calls in the orbit. Our result generalizes earlier results where Poisson input was assumed.

A simulation model enables manufacturers to accurately evaluate the performance of their manufacturing systems. A vital input to build simulation models of manufacturing systems is the system behaviour, which is modelled based on extensive data on manufacturing systems and their elements. However, for outsourced manufacturing systems, data provided from external suppliers to manufacturers are generally limited and insufficient. Therefore, a solution is required to identify and secure system behaviour from available data. We propose herein a behaviour extraction process along with detailed activities and diagrams to extract system behaviour from limited data. By using the systems engineering standard ANSI/EIA-632, the proposed process systematically restructures available data on a manufacturing system and its elements and then provides complete system behaviour to build a simulation model. A case study on a cooling water control system in a hot strip mill convincingly demonstrates the applicability and effectiveness of the proposed process.

Critical infrastructures (CI) are difficult to handle due to their complexity, size and the number of stakeholders involved. During emergency situations (e.g. fire or terrorist attack), a CI operator in the control room is faced with a flood of information coming from different sensors and legacy monitoring systems. Since in these situations, time is critical and the operators are under a great deal of pressure, a holistic management of all the technical systems and actors involved is needed, reinforced by the Recommendation and Decision Support System (RDSS) that helps emergency managers to take correct and timely decisions. One way to provide an adequate RDSS support to the operator is proposed in this paper which is based on an intelligent, event driven layer that sits on top of the legacy CI monitoring system. Powered by the complex event processing capabilities and facility data model implemented in the form of CI ontology, this layer processes events originating from different sources, conducts the situation and risk assessment, and reacts accordingly, either automatically or via recommendations proposed to emergency personnel. To validate the proposed approach, an event-driven RDSS was deployed on an airport use case (Nikola Tesla airport in Belgrade), as one of the most complex CIs.

Uncovering shilling attackers hidden in recommender systems is very crucial to enhance the robustness and trustworthiness of product recommendation. Many shilling attack detection algorithms have been proposed so far, and they exhibit complementary advantage and disadvantage towards various types of attackers. In this paper, we provide a thorough experimental comparison of several well-known detectors, including supervised C4.5 and NB, unsupervised PCA and MDS, semi-supervised HySAD methods, as well as statistical analysis methods. MovieLens 100K is the most widely-used dataset in the realm of shilling attack detection, and thus it is selected as the benchmark dataset. Meanwhile, seven types of shilling attacks generated by average-filling and random-filling model are compared in our experiments. As a result of our analysis, we show clearly causes and essential characteristics insider attackers that might determine the success or failure of different kinds of detectors.

Negotiation is both an important topic in multi-agent systems research and an important aspect of daily life. Many real-world negotiations are complex and involve multiple interdependent issues, therefore, there has been increasing interest in such negotiations. Existing nonlinear automated negotiation protocols have difficulty in finding solutions when the number of issues and agents is large. In automated negotiations covering multiple independent issues, it is useful to separate out the issues and reach separate agreements on each in turn. In this paper, we propose an effective approach to automated negotiations based on recursive partitioning using an issue dendrogram. A mediator first finds partial agreements in each sub-space based on bids from the agents, then combines them to produce the final agreement. When it cannot find a solution, our proposed method recursively decomposes the negotiation sub-problems using an issue dendrogram. In addition, it can improve the quality of agreements by considering previously-found partial consensuses. We also demonstrate experimentally that our protocol generates higher-optimality outcomes with greater scalability than previous methods.