The optimal searching strategies of curbside parking with schedule delay penalties

This paper investigates the drivers’ microscopic curbside parking behaviours under random distribution of vacant parking spaces. For each trip, the driver has a preferred arrival time at the destination and may incur schedule delay cost due to the uncertainty of searching process. The expected trip costs include the in-transit travel cost, the cruising cost, the walking cost, the schedule delay cost and the parking fee. To minimize the expected trip cost, there are two decisions require the driver to make, namely how long before the specified arrival time to initiate the trip and how far away ahead of the destination to search for a vacant parking position. Two possible searching state of the driver are illustrated. The ‘fastest parker’ refers to the possible state of the driver parking successfully when just starts cruising with zero searching distance, and the ‘luckiest parker’ refers to the possible state that driver parks exactly on her/his destination and incurs zero walking distance. Based on the schedule delay states of being the fastest parker and the luckiest parker, three searching strategies are distinguished. The ‘Early-park-early-arrival’ strategy refers to the driver has the expectation that arrives at the destination early when being the fastest parker; the ‘Early-park-late-arrival’ strategy refers to the driver has the expectation that arrives at the destination late when being the fastest parker and arrives early when being the luckiest parker; the ‘Never-early’ strategy refers to the driver has the expectation of late arrival whatever being the fastest or the luckiest parker. The analytical proofs reach some interesting findings: the ‘Never-early’ strategy can never be optimal; if the unit time cost of early arrival is larger than that of late arrival, the optimal parking strategy is ‘Early-park-late-arrival’; if the ‘Early-park-early-arrival’ strategy is optimal, the driver’s possibility of parking before the destination is larger than 50%. Several numerical results are illustrated to verify the analytical propositions and show that the ‘Early-park-early-arrival’ strategy is preferred to the ‘Early-park-late-arrival’ strategy under higher parking charge and higher unit time cost of walking.

Language

  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01601135
  • Record Type: Publication
  • Files: TRIS
  • Created Date: May 18 2016 3:00PM