Predicting project approvals: A case of grants from the National Science Council of Taiwan

Project selection is an important task for organizations in achieving their missions using limited budgets and resources. Whether or not a project will be approved is also of primary concern to the applicants. This paper predicts whether a project will be approved for cases where the criteria for evaluating it are known while the scoring system is not. The idea is to construct a frontier function for the approved projects from past performance on the criteria. The relative distance between a proposed project and the frontier serves as an indicator of the possibility that the project will be approved. Data from the Management II Division of the National Science Council of Taiwan in the Topic Research Project are collected to illustrate this approach. From the percentile of the distance measure, an applicant is able to predict the possibility that their project will be approved. Since professors with different levels of experience and different research areas have different research performance, these factors are taken into account in the prediction. A Malmquist productivity index analysis is also conducted to investigate the performance improvement of the applicants in research between two periods.     

The supplier selection problem with quantity discounts and truckload shipping

To minimize procurement expenditures both purchasing and transportation costs need to be considered. We study a procurement setting in which a company needs to purchase a number of products from a set of suppliers to satisfy customer demand. The suppliers offer total quantity discounts and transportation costs are based on truckload shipping rates. The goal is to select a set of suppliers so as to satisfy product demand at minimal total costs. The resulting optimization problem is strongly NP-hard. We develop integer programming based heuristics to solve the problem. Extensive computational experiments demonstrate the efficacy of the proposed heuristics and provide insight into the impact of instance characteristics on effective procurement strategies.     

Improved Transferability of Data‐Driven Damage Models Through Sample Selection Bias Correction

Damage models for natural hazards are used for decision making on reducing and transferring risk. The damage estimates from these models depend on many variables and their complex sometimes nonlinear relationships with the damage. In recent years, data‐driven modeling techniques have been used to capture those relationships. The available data to build such models are often limited. Therefore, in practice it is usually necessary to transfer models to a different context. In this article, we show that this implies the samples used to build the model are often not fully representative for the situation where they need to be applied on, which leads to a “sample selection bias.” In this article, we enhance data‐driven damage models by applying methods, not previously applied to damage modeling, to correct for this bias before the machine learning (ML) models are trained. We demonstrate this with case studies on flooding in Europe, and typhoon wind damage in the Philippines. Two sample selection bias correction methods from the ML literature are applied and one of these methods is also adjusted to our problem. These three methods are combined with stochastic generation of synthetic damage data. We demonstrate that for both case studies, the sample selection bias correction techniques reduce model errors, especially for the mean bias error this reduction can be larger than 30%. The novel combination with stochastic data generation seems to enhance these techniques. This shows that sample selection bias correction methods are beneficial for damage model transfer.     

基于制造商网络渠道选择的双渠道供应链定价 与服务决策研究

本文考虑由线下零售商实体销售与制造商网络销售构成的网络直销,以及由线下零售商实体销售与线上零售商网络销售构成的网络分销两种制造商双渠道模式,研究“搭便车”行为下制造商模式选择和供应链最优定价与服务决策。研究发现:制造商的渠道选择策略与“搭便车”行为程度、消费者偏好网络渠道程度、价格敏感系数等有关。尤其当消费者受到的服务且对价格的敏感性相同时,制造商会选择网络直销双渠道模式。进一步分析消费者“搭便车”行为和网络渠道偏好对制造商渠道选择和定价决策影响发现,制造商在消费者偏好网络渠道程度比较低时应选择网络分销双渠道模式,而当消费者网络渠道偏好和“搭便车”行为程度都比较高时,制造商应选择网络直销双渠道模式。消费者网络渠道偏好或“搭便车”行为程度越大,两种模式下批发价格应设置的更低。网络直销渠道模式下“搭便车”行为程度越大,网络渠道销售价格应设置的更低,而消费者网络渠道偏好程度越大,网络渠道销售价格应设置得更高。     

具有卖空总量限制的多阶段M—SAD投资组合优化

文章提出具有卖空总量限制、阈值约束和V型交易成本的多阶段均值—半绝对偏差（M-SAD）投资组合优化模型。该模型分别运用均值和半绝对偏衡量资产的收益率和风险。由于交易成本的存在,该模型不满足无后效性的动态优化问题。文章将该模型近似为一般动态规划问题,提出一种新的离散迭代方法,并证明该算法是线性收敛的。最后,文章通过实证研究比较分析卖空总量限制和风险偏好系数取不同值时对投资组合最优策略的影响,验证模型和算法的有效性。     

存在战略购买需求的定价、供给、机制选择和配给研究

本文研究存在战略购买需求的易逝资产销售策略问题。垄断厂商基于利润最大化目标确定易逝资产定价、供给、机制选择和配给策略,战略消费者通过锚定预期价格安排战略购买时机。不同于通常基于效用理论研究定价的思路,本文首先基于锚定效应和跨期价格均衡思想探寻不同战略等待购买规模的市场预期需求曲线和动静态定价区域;其次在众多预期需求曲线中寻找市场有效定价前沿（即有效预期需求曲线）;再次在利润曲面上找出与有效定价前沿对应的容量扩展线（即最大利润曲线）;最后沿容量扩展线和有效定价前沿搜寻最大期望利润及相应策略。研究表明,消费者保留价异质和需求不确定性是动态定价和战略购买存在的根本原因;市场在不同战略等待购买规模状态拥有不同预期需求曲线,最大战略等待购买规模状态预期需求曲线是市场有效定价前沿。动静态定价机制各有其所适用的容量和价格空间,消费者保留价水平和战略消费者规模决定动态定价空间大小,随机需求分布差异只影响动态定价空间形状（即影响需求弹性）。在跨期价格均衡区域内,提价和扩容都会加剧消费者战略购买程度,供给越大定价往往越低。战略购买不仅会降低厂商供给、定价和利润水平,改变不同类型消费者之间高低价购买机会,甚至还可能影响定价机制选择和配给策略。压缩过度供给和虚高价格空间可降低战略购买导致的利润损失。本文研究结果可为考虑消费者行为的需求价格理论研究和运营管理实践提供参考。     

基于SRFML-Lift的流程制造产品质量状态监测

对流程制造型企业而言,产品质量状态的监测精度直接影响了企业的生产与运营成本。面对流程工业的多变量监测要求和数据不均衡性,以往研究主要采取局部建模策略或多输出模型,存在特征选择偏差和分类精度不高的问题。对此,本文设计了一种结合SRFML特征选择和Lift学习策略的质量状态监测模型,通过共享不同目标之间的信息以期提升模型的监测效果。首先,根据ReliefF过滤机制,引入重采样赋权思想对工业特征的选择过程进行优化(SRFML);然后,将选择结果作为Lift学习框架的输入,通过类属属性学习方式重塑各待监测特性的特有关联属性;最后采用多个SVM分类器进行训练,得到各目标的质量状态结果。结果表明,本文构建的SRFML-Lift充分学习了原始特征的关键信息,与其他组合策略相比,对质量状态的监测效果更佳,可应用于流程工业的生产管理实践。     

逆向选择条件下带奖惩金的两期保险契约模型及其Pareto改进性研究

为了更有效的规避影响保险市场交易效率的逆向选择问题,本文分投保人风险类型为两种和多种情形建立了带奖惩金的两期保险契约模型,首次提出可以用奖励金和惩罚金有效甄别投保人的风险类型。该模型根据投保人第一个保险期内的索赔情况在第二个保险期对其进行奖励或惩罚,高风险类型的投保人如果选择为低风险类型投保人设计的保险契约,则其在第二阶段受到惩罚的概率要远远大于得到奖励的概率,即风险越高的投保人越害怕惩罚金,因此所建模型满足斯彭斯-莫里斯分离条件。带奖惩金的两期保险契约模型中保险公司的期望利润仍然为0,并不会给投保人带来额外的经济负担,却能够实现对传统部分保险契约简单重复两次的严格帕累托改进。最后采用一个算例说明了该模型的有效性。     

Fit and Skill in Employee Selection: Insights from a Study of Headhunters

We draw upon evidence from a qualitative study of headhunters to provide insights into the character and importance of candidate fit and skill for the selection of a broad range of white-collar employees. Headhunters suggest that the fit of a job candidate is assessed at two levels, one corresponding with a general compatibility with organization-level norms, culture, and strategy, the other corresponding more closely with traits and characteristics of the person or persons with whom the job candidate actually interviews. Skill—a factor which is largely neglected by those who tout the importance of fit—also plays an important and independent role in employee selection. Stalls that influence the selection of employees from a pool of candidates tend to be highly specific if not idiosyncratic, and take the form of what headhunters call hot buttons. We conclude by discussing the conceptualization, causes, and implications of fit; we also consider how the importance of fit and hot buttons challenges the explanatory logic of standard accounts of labor-market success.     

Teen Childbearing and Educational Attainment: A Comparison of Methods

This research examines the impact of teenage childbearing on secondary school completion, while focusing on the problem of causal ambiguity in the relationships among self-determined behaviours. Techniques for dealing with the teenage childbearing problem are discussed, and results from these methods are compared. Data from the High School and Beyond Study on young women (n = 5257) who were enrolled as sophomores in sample schools in February 1980 and who had not given birth before November 1980 are used. Results indicate that teen childbearing reduces the probability of completing high school by 8% to 10%. Some evidence suggests that programs that target reduction of teen childbearing in improving young women's education and subsequent economic and labor force outcomes.