Network State Estimation and Prediction for Real-Time Traffic Management

Advanced Traveler Information Systems (ATIS) and Advanced Traffic Management Systems (ATMS) have the potential to contribute to the solution of the traffic congestion problem. DynaMIT is a real-time system that can be used to generate guidance for travelers. The main principle on which DynaMIT is based is that information should be consistent, and user optimal. Consistency implies that the traffic conditions experienced by the travelers are consistent with the condition assumed in generating the guidance. To generate consistent user optimal information, DynaMIT performs two main functions: state estimation and prediction. A demand simulator and a supply simulator interact to perform these tasks. A case study demonstrates the value of the system.     

Estimation of Bid Functions for Location Choice and Price Modeling with a Latent Variable Approach

A new approach for the estimation of bid-rent functions for residential location choice is proposed. The method is based on the bid-auction approach and considers that the expected maximum bid of the auction is a latent variable that can be related to observed price indicators through a measurement equation. The method has the advantage of allowing for the estimation of the parameters of the bid function that explain the heterogeneous preferences of households for location while simultaneously adjusting the expected maximum bid to reproduce realistic values. The model is applied and validated for a case study on the city of Brussels. Results show that the proposed model outperforms other methods for bid-rent estimation, both in terms of real estate prices and spatial distribution of agents, especially when detailed data describing the real estate goods and their prices is not available.     

Real-time management of berth allocation with stochastic arrival and handling times

In this research we study the berth allocation problem (BAP) in real time as disruptions occur. In practice, the actual arrival times and handling times of the vessels deviate from their expected or estimated values, which can disrupt the original berthing plan and possibly make it infeasible. We consider a given baseline berthing schedule, and solve the BAP on a rolling planning horizon with the objective to minimize the total realized cost of the updated berthing schedule, as the actual arrival and handling time data is revealed in real time. The uncertainty in the data is modeled by making appropriate assumptions about the probability distributions of the uncertain parameters based on past data. We present an optimization-based recovery algorithm based on set partitioning and a smart greedy algorithm to reassign vessels in the event of disruption. Our research problem derives from the real-world issues faced by the Saqr port, Ras Al Khaimah, UAE, where the berthing plans are regularly disrupted owing to a high degree of uncertainty in information. A simulation study is carried out to assess the solution performance and efficiency of the proposed algorithms, in which the baseline schedule is the solution of the deterministic BAP without accounting for any uncertainty. Results indicate that the proposed reactive approaches can significantly reduce the total realized cost of berthing the vessels as compared to the ongoing practice at the port.     

Modelling human perception of static facial expressions

A recent internet based survey of over 35,000 samples has shown that when different human observers are asked to assign labels to static human facial expressions, different individuals categorize differently the same image. This fact results in a lack of an unique ground-truth, an assumption held by the large majority of existing models for classification. This is especially true for highly ambiguous expressions, especially in the lack of a dynamic context. In this paper we propose to address this shortcoming by the use of discrete choice models (DCM) to describe the choice a human observer is faced to when assigning labels to static facial expressions. Different models of increasing complexity are specified to capture the causal effect between features of an image and its associated expression, using several combinations of different measurements. The sets of measurements we used are largely inspired by FACS but also introduce some new ideas, specific to a static framework. These models are calibrated using maximum likelihood techniques and they are compared with each other using a likelihood ratio test, in order to test for significance in the improvement resulting from adding supplemental features. Through a cross-validation procedure we assess the validity of our approach against overfitting and we provide a comparison with an alternative model based on Neural Networks for benchmark purposes.     

Constraint-specific recovery network for solving airline recovery problems

In this paper, we consider the recovery of an airline schedule after an unforeseen event called disruption, making the planned schedule infeasible. We present a modeling framework that allows the consideration of operational constraints within a Column Generation (CG) scheme. We introduce the general concept of recovery network, generated for each individual unit of the problem, and show how unit-specific constraints are modeled using resources. We fully illustrate the concept by solving the Aircraft Recovery Problem (ARP) with maintenance planning, we give some insights into applying the model to the Passenger Recovery Problem (PRP) and we present computational results on real data.     

Behavioral Priors for Detection and Tracking of Pedestrians in Video Sequences

In this paper we address the problem of detection and tracking of pedestrians in complex scenarios. The inclusion of prior knowledge is more and more crucial in scene analysis to guarantee flexibility and robustness, necessary to have reliability in complex scenes. We aim to combine image processing methods with behavioral models of pedestrian dynamics, calibrated on real data. We introduce Discrete Choice Models (DCM) for pedestrian behavior and we discuss their integration in a detection and tracking context. The obtained results show how it is possible to combine both methodologies to improve the performances of such systems in complex sequences.