Page - 95 - in Emerging Technologies for Electric and Hybrid Vehicles

Energies 2016,9, 86 the routes, involvinga timepenalty. Thisway, several scenarios combiningkindsofvehicleswere evaluated,andfinally theresults showedthatusingEVsisamorecostlyalternativedueto thehigh investmentrequiredforacquiringorconvertingthesevehicles.AccordingtoLebeau et al. (2015) [63], oneof theproblemsof thismodel is that the locationsofchargingspotswerenotconsidered,meaning thatEVscouldvirtuallyrefuelanywhereonthedeliveryroundonce thebatterywasempty. Erdog˘an andMiller-Hooks [65] improvedthepreviousworkbydevelopingthegreenVRPwith thepossibility of refuelingvehiclesat theexistingalternative fuelingstationsalong the routes. Thecontributions of these twoworkswere integratedbySchneider et al. [66] in theirelectricVRPwithtimewindows. Charging locationsandchargingtimesarebothconsidered in theirmodelwhichapproacheswell the problemofEVs. AsimilarapproachwasalsodevelopedbyConradandFigliozzi [67]. Basedona capacitatedVRPwithtimewindowsconstraints, they introducedthe limitedrangeandchargingtimes inorder toget therechargingvehicle routingproblem.Theirmaindifference is regardingcharging locations, since theseauthorsconsider thatcharging ispossibleat somecustomer locationswhile the formulation in [66] ismoreflexibleasotherpossiblecharging locationsarepossible in thenetwork. Bae et al. [68]alsoconsideredtheEVandinternal-combustionenginesvehiclesfleet sizeandmix problemasa two-player two-stagegame. They focusedondeterminationof the levelofhybridor alternativeenergydeliveryfleet fora logisticsandtransportationcompany. Inorder todothat, they constructedamodelof self-selectionwithheterogeneousconsumerswhovalue thefirm’sdelivery servicealongtwodimensions: thequalityofdeliveryserviceandtherelativereduction inemissions. Theseauthorsconcludedthatwhilesubsidiesmayincrease theoperator’sprofit theymayalsoresult inhigherprices for thecustomers.However, thesecustomerswillbenefit fromareduction inpollution ifmoreEVsareused.Morerecently,VanDuin et al. [69]dealtwith theelectricalvehiclefleet sizeand mixVRPwith timewindows. Theaimis todetermineanoptimalfleetofEVsanddelivery routes to offer a desired service level atminimal cost to a set of customerswithdelivery timewindows. However,asnoticedbyLebeau et al. (2015) [63], theyapproachedtheproblemwithoutconsideringthe previousworkonbatteryelectricvehicles inVRP.Asaresult, themodel involvessimilarweaknesses as in [64], i.e., theydonotconsider the locationsofchargingpoints.AnEVwithabatteryswapping systemismodeledso that the rangeof thisEVcanbedoubled.Nonetheless, theswappingsystem isnot reflected in theconstraints. It is in fact reflected in the rangeparameterof thevehiclewhich is simplydoubled,meaning that thebatteryof theEVcanbe swappedvirtually anywhereon the road.Hiermann et al. [70]developedthat idea further toproposeanelectricalvehiclefleet sizeand mixVRPwith timewindowsthatalsoconsiders thedecisionsregardingthefleetcompositionandthe choiceof rechargingtimesandlocations. Thisworkcanbeconsideredas thestateof theartofdelivery optimizationwithEVs. Ifvehiclescannotrechargeorswaptheirbatteriesontheroad, thenanother differentproblemcanbediscussedas in Juan et al. [10],whodealtwith theVRPwithmultipledriving ranges,anextensionof theclassical routingproblemwhere the totaldistanceeachvehiclecantravel is limitedandisnotnecessarily thesameforallvehicles, i.e., thefleet isheterogeneouswithrespect to maximumroute lengths. Regarding hydrogen-basedEVs, additional papers highlight the use of fleet vehicles to take advantageofcentralizedfuelingwhilehydrogen infrastructure isbeingdeveloped. Thus, for instance, Mercuri et al. [71], Joffe et al. [72],O’Garra et al. [73], andBrey et al. [74] present examples in Italy, UnitedKingdom, andSpain, respectively. More recently, researchershavebegun to includemore complete systems into the scope of theirmodels. For example, infrastructuremodels have been developed forChina [75], Europe [76],Germany [77],GreatBritain [78], SouthKorea [79], and the UnitedStates [80–82]. Forarecent literaturereviewaimedatoptimizinghydrogeninfrastructuresee AgnolucciandMcDowall [83]. 4. EmergingVehicleRoutingProblem(VRP)Operational IssuesRelatedto theUseofEVs Novelemergingroutingmodels forEVshaveto includethemost importantpractical constraints of logisticsserviceproviders thatuseEVsfor last-miledeliveries. First,vehiclecapacityrestrictions 95