Major cities—including New York, Paris, and London—have implemented bike-share systems to increase the use of sustainable modes of transportation. In this chapter we discuss the operational challenges of implementing such systems and also provide design recommendations. Critical decisions when implementing bike-share systems are the number of bikes and the number of docking points. For a proposed system, these quantities are determined as a function of the imbalance in demand flow (i.e., of demand asymmetry) in different directions. We show how this asymmetry affects decisions about the number of bikes and docking points and also about the frequency with which bikes should be reallocated from full stations to empty ones. The importance for users of the accessibility to stations and their bike-availability are the key determinants of how many daily trips a system attracts and hence of that system’s optimal design. We describe an empirical study that determines user behavior in a central Parisian bike-share system and then suggest an alternate design which can increase the number of trips by almost 29 % for the same amount of resources in number of bikes based on user behavior parameters that we estimate.
[1]
Nils Rudi,et al.
Newsvendor Networks: Inventory Management and Capacity Investment with Discretionary Activities
,
2002,
Manuf. Serv. Oper. Manag..
[2]
P. Davis.
Spatial competition in retail markets: movie theaters
,
2006
.
[3]
Steven T. Berry,et al.
Automobile Prices in Market Equilibrium
,
1995
.
[4]
David D. Yao,et al.
Fundamentals of Queueing Networks
,
2001
.
[5]
Karan Girotra,et al.
Bike-Share Systems: Accessibility and Availability
,
2018,
Manag. Sci..
[6]
Tal Raviv,et al.
Static repositioning in a bike-sharing system: models and solution approaches
,
2013,
EURO J. Transp. Logist..
[7]
D. Yao,et al.
Fundamentals of Queueing Networks: Performance, Asymptotics, and Optimization
,
2001,
IEEE Transactions on Automatic Control.
[8]
Gérard P. Cachon.
Matching Supply with Demand
,
2000
.
[9]
J. Little.
A Proof for the Queuing Formula: L = λW
,
1961
.