Christian Rudloff

Modeling Demand for Bikesharing Systems: Neighboring Stations as Source for Demand and Reason for Structural Breaks

Bikesharing systems are becoming popular all over the world. One of the remaining problems is that the rides are not uniformly distributed between stations and that certain stations fill up or empty over time. These empty and full stations lead to demand for bikes and return boxes (docks) that cannot be fulfilled; the situation leads to unsatisfied and possibly even lost customers. To avoid this situation, the provider redistributes bikes in the system. Although redistribution of bikes in such systems is well studied, the underlying demand has not yet been modeled to serve as an input to improve the redistribution. For this gap to be closed, demand for bikes and return boxes was modeled with data from the bikesharing system Citybike Wien in Vienna, Austria. In particular, the influence of weather and full or empty neighboring stations on demand was studied by using different count models. Furthermore, historic demand was used to show that forecasts from the model had improved. Last, the influence of new stations on the model parameters of a station and resultant structural breaks in the model were discussed.

Can Walking Behavior Be Predicted? Analysis of Calibration and Fit of Pedestrian Models

Several models for simulation of pedestrian movement have been proposed in recent decades. These models are primarily used in the planning and evaluation of large pedestrian infrastructures, such as transportation hubs, with a focus to increase comfort and safety for pedestrians. Although the number of proposed simulation models is increasing at a fast pace, not much is known about the properties of calibration procedures or the transferability of the models estimated in one setting to other settings. This paper compares three calibration methods for a slightly adapted social force model. The main emphasis lies in the characteristics of the data-generation process and the information contained in the data sets. The sensitivity of the model parameters of the calibrated model were investigated, and the transferability of the model to different scenarios was tested. Results revealed that the quality of the data had a strong effect on the suitability of different calibration strategies and that the information content in the scene under investigation limited the transferability of the results to other scenarios. These results suggest that several data sets with different characteristics do not need to be included in the calibration process to achieve a model that performs well in a wider variety of settings.

Modelling Random Taste Variations on Level Changes in Passenger Route Choice in a Public Transport Station

Abstract In large stations of public transportation high crowd densities can lead to potential safety risks and to unnecessary delays. To assess the actual capacity of potential bottlenecks a deeper understanding on the route choice of pedestrians is of great importance. This paper investigates the factors that influence the route choice of pedestrians when facing a stair/escalator combination in a major Austrian train station. We employ random utility models on data sets of revealed and stated preferences. In particular we investigate the potential for heterogeneities in taste by employing mixed logit models. The results show that, first, crowding is an important factor for route choice, second, that the application of mixed logit models is appropriate and, last, that the use of both revealed and stated preference data adds valuable information.

Integrating inventory control and a price change in the presence of reference price effects: a two-period model

Demand and procurement planning for consumer electronics products must cope with short life cycles, limited replenishment opportunities and a willingness to pay that is influenced by past prices and decreases over time. We therefore propose the use of an integrated pricing and inventory control model with a two-period linear demand model, in which demand also depends on the difference between a price-history-based reference price and the current price. For this model we prove that the optimal joint pricing/inventory policy for the replenishment opportunity after the first period is a base-stock list-price policy. That is, stock is either replenished up to a base-stock level and a list-price is charged, or it is not replenished and a discount is given that increases with the stock-level. Furthermore, we use real-world cell phone data to study the differences between an integrated policy and traditional sequential optimization, where prices are initially optimized based on the expected demand and ordering cost, and the resulting demand distribution is used to determine an optimal inventory policy. Finally, we discuss possible extensions of the model.

Personalized and Situation-Aware Multimodal Route Recommendations: The FAVOUR Algorithm

Route choice in multimodal networks shows a considerable variation between different individuals and the current situational context. Personalization and situation awareness of recommendation algorithms are already common in many areas, e.g., online retail. However, most online routing applications still provide shortest distance or shortest travel-time routes only, neglecting individual preferences, as well as the current situation. Both aspects are of particular importance in a multimodal setting as attractivity of some transportation modes, such as biking, which crucially depend on personal characteristics and exogenous factors, such as the weather. As an alternative, this paper introduces the FAVourite rOUte Recommendation (FAVOUR) approach to provide personalized, situation-aware route proposals based on three steps: first, at the initialization stage, the user provides limited information (home location, work place, mobility options, sociodemographics) used to select one out of a small number of initial profiles. Second, based on this information, a stated preference survey is designed in order to sharpen the profile. In this step, a mass preference prior (MPP) is used to encode the prior knowledge on preferences from the class identified in step one. Third, subsequently, the profile is continuously updated during usage of the routing services. The last two steps use Bayesian learning techniques in order to incorporate information from all contributing individuals. The FAVOUR approach is presented in detail and tested on a small number of survey participants. The experimental results on this real-world dataset show that FAVOUR generates better quality recommendations w.r.t. alternative learning algorithms from the literature. In particular, the definition of the MPP for initialization of step two is shown to provide better predictions than a number of alternatives from the literature.

Daily travel behavior: lessons from a week-long survey for the extraction of human mobility motifs related information

Multi-agent models for simulating the mobility behavior of the urban population are gaining momentum due to increasing computing power. Such models pose high demands in terms of input data in order to be reliably able to match real world behavior. To run the models a synthetic population mirroring typical mobility demand needs to be generated based on real world observations. Traditionally this is done using travel diary surveys, which are costly (and hence have relatively low sample size) and focus mainly on trip choice rather than on activities for an entire day. Thus in this setting the generation of synthetic populations either relies on resampling identical activity chains or on imposing independence of various trips occurring during the day. Both assumptions are not realistic. Using Call Detail Records (CDRs) it has been found that individual daily movement uses only a small number of movement patterns. These patterns, termed motifs, appear stably in many different cities, as has been shown for both CDR data as well as travel diaries. In this paper the relation between these motifs and other mobility related quantities like the distribution of travel distances and times as well as mode choice is investigated. Additionally transition probabilities both for motifs (relevant for multi-day simulations) and mode transitions are discussed. The main finding is that while some of the characteristics seem to be unrelated to motifs, others such as mode choice exhibit strong correlations which could improve the provision of synthetic populations for multi-agent models. Thus the results in this paper are seen as one step further towards the creation of realistic (with respect to mobility behavior) synthetic populations for multi-agent models in order to analyze the performance of multi-modal transportation systems or disease spreading in urban areas.

Comparison of Different Calibration Techniques on Simulated Data

Pedestrian simulation models are used in a multitude of ways. While model development for Social Force models is quite advanced little is known about how to calibrate a Social Force model such that the parameters are reproducing real life data in many situations. This paper tests different calibration methodologies that were previously used in the calibration of Social Force models to give an indication of the usability of such techniques. This is done on simulated data to guarantee the comparability of the different methodologies. The result will give practitioners a guideline in choosing how to use real live data in the calibration process.

Comparing Calibrated Shared Space Simulation Model with Real-Life Data

Shared spaces are being implemented in many countries to deal with safety concerns and traffic flow problems on busy urban streets and street crossings. However, shared space concepts could not be tested before they were built because of the lack of a functioning microscopic shared space simulation. Lane-based car-following models, currently used in traffic simulation, cannot reproduce the high heterogeneity of a mixed traffic modes nonchannelized flow. This papers novel approach introduces an extended social force model for vehicles and pedestrians that incorporates social interactions between different modes of transport rather than following a purely rule-based approach. The calibration of such a microscopic traffic simulation model with real-world data from two shared space sites is presented. The simulation can reproduce real-life shared space behavior by comparing it with trajectory and interaction data collected at implemented shared space road designs.

Travel Behavior and Electric Mobility in Germany: Is the Problem the Driving Range, Costs, or Both?

A reluctance to switch to electric vehicles is observed in various countries despite national efforts to promote them. The question of whether electric cars are capable of meeting daily mobility requirements in Germany is investigated. The analysis is based on data from the German Mobility Panel Survey from 1995 to 2010 and the travel survey of 2009 and 2010 for the Stuttgart, Germany, area and combines a long-term travel behavior analysis with a region-specific verification. The focus is on individuals who exclusively drive a car and walk during the day and who rely on the car as a primary means of transport. For this group, the determinants of the decision to drive an internal combustion engine vehicle versus a battery electric vehicle (BEV) are analyzed, with a focus on driving range and energy costs. Results of the analysis suggest that around 80% of all daily travel by car drivers could be done with currently available models of electric cars and that charging them only at night would be sufficient in most cases. Therefore, the driving range of BEVs cannot be the restricting factor. In contrast, the current cost structure of BEVs (high investment cost, low energy cost) is not favorable for the large share of drivers with low annual mileage because the high investment cost is not compensated for by low operation costs. In the Stuttgart region, drivers from the suburbs would benefit most from such energy cost savings; however, city dwellers would need other cost structures or incentives to switch to BEVs.

Mind the gap: Boarding and alighting processes using the social force paradigm calibrated on experimental data

One of the few possibilities to increase the capacity of train lines is to make boarding and alighting more efficient. So far expensive experiments were needed to evaluate the effect of proposed changes to the train setup. This paper shows that a simulation model based on the social force paradigm and calibrated on measured data promises that in future simulations could be used to test new scenarios in a cheap and fast way.