Overview and details of the sessions of this conference. Please select a date or location to show only sessions at that day or location. Please select a single session for detailed view (with abstracts and downloads if available).
D2S1T3: Transportation Networks and Vehicle Routing
11:30am - 12:30pm
Session Chair: Daniel Schmand
Simulative Assessment of Patrol Car Allocation and Response Time
Cors, Tobias; Fliedner, Malte
Universität Hamburg, Germany
In the planning of police forces an adequate allocation of resources to districts is crucial for providing an effective and robust service and for maintaining acceptable emergency response times. Due to the high operational heterogeneity and variability, determining reliable profiles for resource utilization and establishing their relationship to response times is a challenging task in and of itself that requires an adequate consideration of several sources of stochastic influence. Prior approaches from police practice mainly consider static ratios (e.g. resources per number of inhabitants or calls for service) in order to estimate capacity demand. Based on an extensive dataset comprising more than two million items, we estimate stochastic process models for all relevant police operations in a major metropolitan area and use a discrete-event simulation to analyze capacity utilization and response times of a given fleet of police vehicles. The simulation model predicts the spatial and temporal occurrence of police operations and dispatches available vehicles from different districts, in order to model resource sharing in emergency response. The human dispatcher is mimicked by an observed set of static and workload-based rules. The vehicle allocation is then optimized via a rule-based heuristic, observing shift patterns and local demand structure, until response time expectations are met. This provides key insights into the required capacity over time and constitutes a crucial first step for an adequate capacity planning.
A new lower bound for the static dial-a-ride problem with ride and waiting time minimization
Pfeiffer, Christian; Schulz, Arne
Institute of Operations Management, University of Hamburg, Germany
The paper focuses on the static dial-a-ride problem with ride and waiting time minimization. This is an important problem setting of significant practical relevance, as several ridesharing providers launched in recent years in large cities. In contrast to the standard dial-a-ride problem, these providers focus on the general public. Therefore, they are amongst others in competition with taxis and private cars, which makes a more customer-oriented objective necessary. We minimize the sum of relative detours of all customers. The paper introduces upper bounds for the arrival times and an initial lower bound for the objective value. Our approach is tested in a computational study with realistic test instances.
A branch-and-cut algorithm for the dial-a-ride problem with ride and waiting time minimization and time windows
Schulz, Arne; Pfeiffer, Christian
Universität Hamburg, Institut für Operations Management, Hamburg, Germany
In recent years, several ridesharing providers launched in large cities. In contrast to the standard dial-a-ride problem, these providers focus on the general public. As they are amongst others in competition with taxis and private cars, it is important for them to ensure short travel times for the customers. In this talk, we consider the resulting dial-a-ride problem with ride and waiting time minimization, which minimizes the relative detours of all customers in relation to the earliest point in time the customer can be picked up. Furthermore, we assume time windows for every customer request. We present a branch-and-cut algorithm with several techniques to improve the search. Moreover, we analyse our techniques in a computational study.