Bayes Inference in Constant Partially Accelerated Life Tests for the Generalized Exponential Distribution with Progressive Censoring

In this paper, the problem of constant partially accelerated life tests when the lifetime follows the generalized exponential distribution is considered. Based on progressive type-II censoring scheme, the maximum likelihood and Bayes methods of estimation are used for estimating the distribution parameters and acceleration factor. A Monte Carlo simulation study is carried out to examine the performance of the obtained estimates.     

Updating discriminant functions estimated from inverse gaussian populations

The problem of updating discriminant functions estimated from inverse Gaussian populations is investigated in situations when the additional observations are mixed (unclassified) or classified. In each case two types of discriminant functions, linear and quadratic, are considered. Using simulation experiments the performance of the updating procedures is evaluated by means of relative efficiencies.     

Estimation of odds ratios under order restrictions

A new method for estimating a set of odds ratios under an order restriction based on estimating equations is proposed. The method is applied to those of the conditional maximum likelihood estimators and the Mantel-Haenszel estimators. The estimators derived from the conditional likelihood estimating equations are shown to maximize the conditional likelihoods. It is also seen that the restricted estimators converge almost surely to the respective odds ratios when the respective sample sizes become large regularly. The restricted estimators are compared with the unrestricted maximum likelihood estimators by a Monte Carlo simulation. The simulation studies show that the restricted estimates improve the mean squared errors remarkably, while the Mantel-Haenszel type estimates are competitive with the conditional maximum likelihood estimates, being slightly worse.     

A hybrid approach to configure eco-efficient supply chains under consideration of performance and risk aspects

Formal models that support multi-criteria decision making represent a strongly growing area in sustainable supply chain management research. However, uncertainties and risks are seldom considered in quantitative models for green supply chain (SC) design. The paper at hand suggests a hybrid approach to configure an eco-efficient SC for a new product under consideration of economic and environmental risks. Discrete-event simulation is applied to assess the financial, operational and environmental performance of different SC configuration options while the value-at-risk concept is adapted to evaluate related SC risks. The analytic hierarchy process is employed to solve the resulting multi-criteria decision problem of choosing the best option. The approach is illustrated at a case example of a fast moving consumer goods manufacturer.     

考虑合作程度的电信数据业务合作创新模型研究

基于模型试验，对堆载诱发型边坡滑坡的变形机理及演化过程进行研究。研究结果表明: （1）堆载诱发型滑坡的变形演化规律可归纳为：后缘压缩阶段→蠕动变形阶段→加速滑动阶段→剧滑阶段，且剧滑启动之前的加速变形过渡时间极短；（2）利用FLAC3D软件进行堆载滑坡演化过程的动态数值分析，模拟过程中坡体应力场和位移场的演化特征表现为从上向下逐步贯通，与模型边坡破坏过程相符；（3）堆载诱发型滑坡破坏过程历时短、突发性强，在坡脚产生持续位移时即应做出滑坡预警。     

强化学习型匝道控制模型的研究

高速公路上的交通堵塞造成了道路利用效率低下，并伴随着能源消耗和环境污染问题，因此各种各样的高速公路控制方法应用于缓解交通堵塞。本文提出强化学习型匝道控制模型，该模型以交通流模拟为预测工具，以人工智能的强化学习为最优化选择模型，并具有一定的自主性、有记忆功能和性能反馈功能，且是一种动态的过程。应用JAVA针对不同的交通状态进行模拟再现，模拟结果表明匝道控制模型对于减少交通堵塞具有显著的效果。     

Interval estimation for misclassification rate parameters in a complementary Poisson model

We investigate three interval estimators for binomial misclassification rates in a complementary Poisson model where the data are possibly misclassified: a Wald-based interval, a score-based interval, and an interval based on the profile log-likelihood statistic. We investigate the coverage and average width properties of these intervals via a simulation study. For small Poisson counts and small misclassification rates, the intervals can perform poorly in terms of coverage. The profile log-likelihood confidence interval (CI) is often proved to outperform the other intervals with good coverage and width properties. Lastly, we apply the CIs to a real data set involving traffic accident data that contain misclassified counts.     

Confidence intervals on p(y < x) for small sample sizes

When F = G^a, confidence intervals are derived and presented in graphs for p = P(Y < X), when X and Y are independent and the sample sizes are at most 25. Also, it is demonstrated via a monte carlo simulation that this is a robust procedure, when the distributions of X and Y differ by a location parameter.     

Study of the behaviour of flexible production systems using neural nets

Owing to high development and acquisition costs, production systems require very detailed analysis for pre-design specifications, and an adjustment in operating philosophies. Detailed simulation models are one of the most important instruments to achieve this goal; normally the simulator must be custom-built, based on an accurate study of the specific industrial problem. The proposed management and economical analysis uses mathematical meta-models obtained from experimental designs produced by the simulator; today the detail level and precision required is very high, so these kinds of meta-model operate very close to their limit. Here we present a new methodology based on artificial neural networks that produces new types of meta-models which can correctly represent complex industrial systems. In this work we propose an application of this methodology together with a detailed analysis of the model development phase to highlight the potential of this instrument.     

Push-Kanban – a kanban-based production control concept for job shops

Job-shop production in an individual and small-batch manufacturing environment demands producing simultaneously different products in many customised versions and in small numbers. The associated variability of the time-phased work content leads to a high degree of complexity and turbulence in the manufacturing process of such a production type. Therefore, production control becomes significantly more complicated, and many theories on how to cope fail in the real world. Despite the many technological and conceptual advances in the field of job-shop control over the past few decades, there is still a remarkable lack of practical control approaches for job-shop production. For this reason the Push-Kanban production control concept will be introduced in this article. It combines robust push scheduling along with a kanban-driven, decentralised inventory control mechanism to create a holistic control approach, integrating all production control tasks. Thus, it offers a feasible option for controlling job-shop production in industrial practice. This article examines the effectiveness of the concept by means of a simulation study based on real data.