TY - JOUR
T1 - Data-driven Linear Decision Rule Approach for Distributionally Robust Optimization of On-line Signal Control
AU - Liu, Hongcheng
AU - Han, Ke
AU - Gayah, Vikash
AU - Friesz, Terry
AU - Yao, Tao
N1 - Funding Information:
This case study uses data collected within the CARBOTRAF project, which is funded by the 7th EU Framework (http://www.carbotraf.eu). The historical traffic flow data are obtained from Sky High CountOnUs, a company that provides standard turn-by-turn traffic counts for all the signalized intersections in the test network throughout the years 2007-2009. Since traffic counts collected at various locations were on different dates, these data were converted to the same reference date (June 7, 2010) using scaling factors to avoid biased estimation due to temporal effects on traffic. The scaling factors are derived from the historical Annual Average Daily Flow data at the west end of Glasgow (Transport for Scotland, http://www.transportscotland.gov.uk/map-application), and take into account two sources of flow variation: the seasonality effect and the day-of-the-week effect. The link-specific data are extracted from a coordinated use of map data (UK Ordnance Survey, http://www.ordnancesurvey.co.uk/) and microsimulation conducted within the CARBOTRAF project. The time-dependent vehicle turning percentages are derived from the turn-by-turn traffic counts, which record flows associated with various vehicle movements through an intersection. Our study period spans one hour during the morning peak (8:00-9:00am, June 7, 2010).
Publisher Copyright:
© 2015 The Authors. © 2015 The Authors. Published by Elsevier B.V.
PY - 2015
Y1 - 2015
N2 - We propose a two-stage, on-line signal control strategy for dynamic networks using a linear decision rule (LDR) approach and a distributionally robust optimization (DRO) technique. The first (off-line) stage formulates a LDR that maps real-time traffic data to optimal signal control policies. A DRO problem is then solved to optimize the on-line performance of the LDR in the presence of uncertainties associated with the observed traffic states and ambiguity in their underlying distribution functions. We employ a data- driven calibration of the uncertainty set, which takes into account historical traffic data. The second (on-line) stage implements a very efficient linear decision rule whose performance is guaranteed by the off-line computation. We test the proposed signal control procedure in a simulation environment that is informed by actual traffic data obtained in Glasgow, and demonstrate its full potential in on-line operation and deployability on realistic networks, as well as its effectiveness in improving traffic.
AB - We propose a two-stage, on-line signal control strategy for dynamic networks using a linear decision rule (LDR) approach and a distributionally robust optimization (DRO) technique. The first (off-line) stage formulates a LDR that maps real-time traffic data to optimal signal control policies. A DRO problem is then solved to optimize the on-line performance of the LDR in the presence of uncertainties associated with the observed traffic states and ambiguity in their underlying distribution functions. We employ a data- driven calibration of the uncertainty set, which takes into account historical traffic data. The second (on-line) stage implements a very efficient linear decision rule whose performance is guaranteed by the off-line computation. We test the proposed signal control procedure in a simulation environment that is informed by actual traffic data obtained in Glasgow, and demonstrate its full potential in on-line operation and deployability on realistic networks, as well as its effectiveness in improving traffic.
UR - http://www.scopus.com/inward/record.url?scp=84959342449&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84959342449&partnerID=8YFLogxK
U2 - 10.1016/j.trpro.2015.06.028
DO - 10.1016/j.trpro.2015.06.028
M3 - Article
AN - SCOPUS:84959342449
VL - 7
SP - 536
EP - 555
JO - Transportation Research Procedia
JF - Transportation Research Procedia
SN - 2352-1457
ER -