With the continuous improvement and observable benefits of electric vehicles (EVs), major logisti... more With the continuous improvement and observable benefits of electric vehicles (EVs), major logistic companies are introducing more EVs into their conventional fleets. This gives rise to a new type of vehicle routing problem with mixed vehicles, where heterogeneous internal combustion vehicles (ICVs) and electric vehicles are considered in route planning. In addition, certain deliveries that are not efficient on any type of vehicles, are outsourced to third-party common carriers. In this paper, we define this problem as a mixed vehicle routing problem with common carriers (MVRPC). The objective of such problems is to minimize the transportation costs by considering routes with ICVs and EVs, the possibility of visiting recharging stations, outsourcing options, and drivers’ layover regulations. This variant of the vehicle routing problems has many practical applications, particularly in the design of long-haul transportation and last-mile delivery services. Effective MVRPC solutions play a key role in promoting the going Green image and optimally allocating resources. The problem has received limited attention in the literature likely because addressing all the needed aspects is especially challenging. To solve the large-scale problem, we develop a branch-and-cut pricing framework that relies on strong cuts and customized labeling algorithms. Numerical experiments highlight the effectiveness of our algorithm. This success can be attributed to tailored critical resources, dynamically bounded bidirectional labeling procedures, strong dominance criteria, and implementation strategies.
Communications for Statistical Applications and Methods, 2018
Risk management has been a crucial part of the daily operations of the financial industry over th... more Risk management has been a crucial part of the daily operations of the financial industry over the past two decades. Value at Risk (VaR), a quantitative measure introduced by JP Morgan in 1995, is the most popular and simplest quantitative measure of risk. VaR has been widely applied to the risk evaluation over all types of financial activities, including portfolio management and asset allocation. This paper uses the implementations of multivariate GARCH models and copula methods to illustrate the performance of a one-day-ahead VaR prediction modeling process for high-dimensional portfolios. Many factors, such as the interaction among included assets, are included in the modeling process. Additionally, empirical data analyses and backtesting results are demonstrated through a rolling analysis, which help capture the instability of parameter estimates. We find that our way of modeling is relatively robust and flexible.
DHL Supply Chain North America moves more than 20 million packages each year. DHL transportation ... more DHL Supply Chain North America moves more than 20 million packages each year. DHL transportation planners perform routing and cost-deduction tasks for many business projects. We refer to the associated planning problem as the Vehicle Routing Problem with Time Regulations and Common Carriers (VRPTRCC). Unlike ordinary vehicle routing problems, which use only a single type of transportation mode, our VRPTRCC applications include make–buy decisions because some of the package deliveries are ultimately subcontracted to organizations other than DHL. Time regulation means that the problem considers not only delivery-time windows, but also layover and driving-time restrictions. Our developed Network Mode Optimization Tool (NMOT) is an ant-colony optimization (ACO)-based program that aids DHL Supply Chain transportation analysts in identifying cost savings in the ground logistic network. By using the NMOT, DHL and its customers have saved millions of dollars annually. Also, the NMOT is help...
Questions of trust in machine-learning models are becoming increasingly important as these tools ... more Questions of trust in machine-learning models are becoming increasingly important as these tools are starting to be used widely for high-stakes decisions in medicine and criminal justice. Transparency of models is a key aspect affecting trust. This paper reveals that there is new technology to build transparent machine-learning models that are often as accurate as black-box machine-learning models. These methods have already had an impact in medicine and criminal justice. This work calls into question the overall need for black-box models in these applications.
With the continuous improvement and observable benefits of electric vehicles (EVs), major logisti... more With the continuous improvement and observable benefits of electric vehicles (EVs), major logistic companies are introducing more EVs into their conventional fleets. This gives rise to a new type of vehicle routing problem with mixed vehicles, where heterogeneous internal combustion vehicles (ICVs) and electric vehicles are considered in route planning. In addition, certain deliveries that are not efficient on any type of vehicles, are outsourced to third-party common carriers. In this paper, we define this problem as a mixed vehicle routing problem with common carriers (MVRPC). The objective of such problems is to minimize the transportation costs by considering routes with ICVs and EVs, the possibility of visiting recharging stations, outsourcing options, and drivers’ layover regulations. This variant of the vehicle routing problems has many practical applications, particularly in the design of long-haul transportation and last-mile delivery services. Effective MVRPC solutions play a key role in promoting the going Green image and optimally allocating resources. The problem has received limited attention in the literature likely because addressing all the needed aspects is especially challenging. To solve the large-scale problem, we develop a branch-and-cut pricing framework that relies on strong cuts and customized labeling algorithms. Numerical experiments highlight the effectiveness of our algorithm. This success can be attributed to tailored critical resources, dynamically bounded bidirectional labeling procedures, strong dominance criteria, and implementation strategies.
Communications for Statistical Applications and Methods, 2018
Risk management has been a crucial part of the daily operations of the financial industry over th... more Risk management has been a crucial part of the daily operations of the financial industry over the past two decades. Value at Risk (VaR), a quantitative measure introduced by JP Morgan in 1995, is the most popular and simplest quantitative measure of risk. VaR has been widely applied to the risk evaluation over all types of financial activities, including portfolio management and asset allocation. This paper uses the implementations of multivariate GARCH models and copula methods to illustrate the performance of a one-day-ahead VaR prediction modeling process for high-dimensional portfolios. Many factors, such as the interaction among included assets, are included in the modeling process. Additionally, empirical data analyses and backtesting results are demonstrated through a rolling analysis, which help capture the instability of parameter estimates. We find that our way of modeling is relatively robust and flexible.
DHL Supply Chain North America moves more than 20 million packages each year. DHL transportation ... more DHL Supply Chain North America moves more than 20 million packages each year. DHL transportation planners perform routing and cost-deduction tasks for many business projects. We refer to the associated planning problem as the Vehicle Routing Problem with Time Regulations and Common Carriers (VRPTRCC). Unlike ordinary vehicle routing problems, which use only a single type of transportation mode, our VRPTRCC applications include make–buy decisions because some of the package deliveries are ultimately subcontracted to organizations other than DHL. Time regulation means that the problem considers not only delivery-time windows, but also layover and driving-time restrictions. Our developed Network Mode Optimization Tool (NMOT) is an ant-colony optimization (ACO)-based program that aids DHL Supply Chain transportation analysts in identifying cost savings in the ground logistic network. By using the NMOT, DHL and its customers have saved millions of dollars annually. Also, the NMOT is help...
Questions of trust in machine-learning models are becoming increasingly important as these tools ... more Questions of trust in machine-learning models are becoming increasingly important as these tools are starting to be used widely for high-stakes decisions in medicine and criminal justice. Transparency of models is a key aspect affecting trust. This paper reveals that there is new technology to build transparent machine-learning models that are often as accurate as black-box machine-learning models. These methods have already had an impact in medicine and criminal justice. This work calls into question the overall need for black-box models in these applications.
Uploads
Papers by Yibo Dang