Parking Lot Management for Charging Stations

Sevket Gökay, Christoph Terwelp, Christian Samsel,

Karl-Heinz Krempels, Sebastian Rabenhorst and Bastian Greber

RWTH Aachen University, Informatik 5, Ahornstr. 55, 52074 Aachen, Germany

Keywords:

Electric Vehicle, Charging Station, Parking Management, Open Charge Point Protocol.

Abstract:

During the last years electric vehicles started to gain signiﬁcant attention from customers and car manufac-

turers. As the number of electric vehicles on the road increases, new requirements emerge to update existing

charging station infrastructures. In this paper we address the issue of management of parking lots where charg-

ing stations are deployed since existing parking management systems do not consider charging stations and

charging station management systems are not able to manage parking spaces. We propose a solution in which

a charging station uses sensors to detect whether the assigned charging spaces are occupied; monitors cus-

tomer’s interaction with itself and forces to take actions within the given period of time. If there is a violation

of terms of use, our solution enables to inform the management system in order to take the necessary steps.

Furthermore, it provides visual guidance to customers in search for parking space and during the interaction.

1 INTRODUCTION

In recent years, there has been a growing interest

in electric vehicles due to concerns about climate

change, greenhouse gas emissions and fossil fuel

prices. With technological innovations in battery and

power management many car manufacturers embrace

producing electric vehicles. Despite the fact that more

charging stations are deployed every year in park-

ing lots, their numbers are still limited. As a result

the more efﬁcient use of parking spaces where public

charging stations are deployed becomes invaluable:

These parking spaces should only be occupied by au-

thorized electric vehicles. Yet existing parking man-

agement systems and solutions only address the avail-

ability of free parking spaces.

In this paper we propose a solution that can be

employed in existing charging stations. The con-

cept involves the use of sensors in order to deter-

mine whether a vehicle is parking or not, monitor-

ing customer interaction with the charging station

while introducing time-based restrictions and warn-

ing him/her accordingly. If the warning is ignored and

therefore the use of the charging station violated, we

provide a mechanism to inform a central system about

the situation. Moreover, this management system is

coupled with a guidance system that proposes the use

of luminaries with three basic colors. In the follow-

ing we differentiate between the terms parking lot and

parking space. A parking space is a location for park-

ing of one vehicle. A parking lot is a dedicated area

for parking of multiple vehicles and thereby consist-

ing of many parking spaces.

The remainder of this paper is structured as fol-

lows. Section 2 describes related work and Section

3 provides the fundamentals of the proposed solution

on an abstract level. Section 4 describes the required

changes considering an existing protocol. Section 5

gives an overview of the implementation and Section

6 concludes.

2 RELATED WORK

A number of the works about charging stations fo-

cus on location planning and charge scheduling.

These are particularly important for distributing load

and therefore increasing efﬁciency. (Hanabusa and

Horiguchi, 2011) and (Hess et al., 2012) propose op-

timizations for charging station placement while tak-

ing distance to travel and waiting times into account.

(Mehar and Senouci, 2013) deals with the same prob-

lem but considers additional factors to minimize de-

ployment costs.

The work in (Timpner and Wolf, 2012) evaluates

various scheduling algorithms for efﬁcient utilization

of charging stations in project V-Charge. Parking

spaces with and without charging stations are man-

289

Gökay S., Terwelp C., Samsel C., Krempels K., Rabenhorst S. and Greber B..

Parking Lot Management for Charging Stations.

DOI: 10.5220/0004879202890295

In Proceedings of the 3rd International Conference on Smart Grids and Green IT Systems (SMARTGREENS-2014), pages 289-295

ISBN: 978-989-758-025-3

 2014 SCITEPRESS (Science and Technology Publications, Lda.)

aged by a future system where vehicles autonomously

navigate as needed regarding the battery’s state of

charge. (Ruzmetov et al., 2013) proposes a platform

where an electric vehicle is automatically assigned to

a charging station based on parameters like the power

level of the battery, distance to travel, and status of

the road trafﬁc. Moreover, the driver is guided to this

charging station with a mobile application.

Open Charge Point Protocol (OCPP) 1.5 (OCPP

Steering Group, 2012) is a standard open protocol

for communication between charging stations and

a central system that manages them. It deﬁnes

message types for authorization of customers, start-

ing/stopping charging of electric vehicles, conﬁgura-

tion and reservation of charging stations. However,

it lacks a mechanism for the management of parking

spaces assigned to the charging station. On the other

hand, (U.S. DOT, 2011) and (Litman, 2013) summa-

rize current solutions in parking management but do

not consider parking lots with charging stations.

3 APPROACH

This section describes our approach on an abstract

level. The charging station is equipped with an occu-

pancy sensor for each single parking space and each

parking space is assigned to a socket of the charg-

ing station. Charging stations are connected to a cen-

tral system. We modeled the contexts of a charg-

ing station as a set of states – free, occupied, au-

thorized, charging, warning, violated – depending on

three sources:

1. Sensor Information: Occupancy sensors can de-

tect whether a parking space is free (no vehicle

parking) or occupied (vehicle parking). Subse-

quently, the central system can be informed about

the change.

2. Customer Interaction with the Charging Sta-

tion: After a parking space is occupied, the cus-

tomer must be authorized (allowed to charge) and

start charging his/her electric vehicle (for exam-

ple by plugging in the cable).

We also added the state warning which implies

that the vehicle has to leave the parking space:

I. The charging station changes from occupied

to the warning state, if the authorization

fails.

II. The charging station changes from charging

to the warning state, if the charging process

is stopped (for example by unplugging the

cable).

3. Timeouts: In order to prevent misuses or abuses

of the parking space we introduced time con-

straints for three states, occupied, authorized and

warning. While a vehicle is parking, the customer

must take actions within a ﬁxed period of time so

that the parking space is not blocked for a long

time and the charging station can offer its service

to others.

I. In the occupied state, the customer must

authorize himself/herself within a ﬁxed pe-

riod of time, otherwise the charging station

changes to the warning state.

II. In the authorized state, the customer must

start charging within a ﬁxed period of time,

otherwise the charging station changes to the

warning state.

III. In the warning state, the vehicle has to leave

the parking space within a ﬁxed period of

time, otherwise the charging station changes

to the violated state. Changing to the vio-

lated state is the only transition where a net-

work communication is essential since cen-

tral system must be informed about the sit-

uation and additional steps should be taken

(such as reporting to law enforcement ofﬁ-

cials).

Timeouts are handled by each charging station in-

ternally, so the task will not be burden on the cen-

tral system.

Figure 1 illustrates the functionality of the charging

station with the states. It is important to note that state

changes are managed by the charging station. This

enables to continue using the functionality if the net-

work connection fails and the charging station is not

connected to the central system. Beside the obvious

hardware-enhancement of including occupancy sen-

sors, the solution proposes changes on the software

of both the charging station and the central system.

First, the charging station must be able to process in-

formation gathered by sensors and change states ac-

cordingly. Second, the charging station’s internal pro-

cesses must trigger state changes when authorization

succeeds or fails, and when charging process starts

and stops. Third, the functionality suggests a commu-

nication between the charging station and the central

system.

Authorization. We made no distinction between

different authorization mechanisms since the chosen

mechanism does not affect the way state changes

function. They all serve the purpose of allowing the

customer to charge or not, hence trigger a transition

SMARTGREENS2014-3rdInternationalConferenceonSmartGridsandGreenITSystems

290

authorized

warningoccupied

charging

free violated

vehicle

parks

authorized

A. authorization failed

B. timeout

vehicle

leaves

vehicle

leaves

charging

starts

charging

stops

timeout

vehicle

leaves

vehicle leaves

timeout

Figure 1: The state diagram of the charging station.

from occupied to authorized or to warning. Differ-

ent operators may use different mechanisms such as

phone calls, sending text messages, near ﬁeld commu-

nication (NFC) with smartphones or radio-frequency

identiﬁcation (RFID) tags attached to membership

cards, charging cables or even vehicles. In all mecha-

nisms the customer is typically given an ID and during

an authorization process it is checked whether the ID

is registered in the database.

Reservation. We also made no exception for the

reservation of a charging station for a speciﬁc cus-

tomer. The reasoning is as follows: If a charging sta-

tion socket is reserved for a speciﬁc ID, then only this

ID will be authorized to charge and authorization of

other IDs will fail even though they might be regis-

tered. On the other hand, if there is no reservation at

that time, all registered IDs can authorize. The logical

distinction between these two cases does not reﬂect on

the implementation of state changes.

3.1 Color Codes

Each parking space features luminaries that code the

states (Table 1). The color codes can be divided into

two groups: Colors when no vehicle is parking and

colors when a vehicle is parking. The ﬁrst group of

colors are intended to guide a customer in search of

a parking space from a distance. The second group

of colors serve the purpose of guiding the customer

during the service of a charging station.

The free state has the extra conditions ready, re-

Table 1: Color codes.

State Color

free & ready green

free & reserved yellow

free & inoperative red

occupied blink green

authorized blink yellow

charging yellow

warning blink red

violated red

served and inoperative. These can be set both by the

charging station and the central system. If a charging

station is ready, it is functional and available to charge

for any one. If a charging station is inoperative, it is

not functional and therefore not available for charg-

ing. If a charging station is reserved, there exists a

reservation for a speciﬁc customer.

The choice of the color scheme based on three cri-

teria: The number of colors should be kept at min-

imum in order to not confuse the customer. Sec-

ond, the colors should be distinct from each other

so that the customer can recognize them from a dis-

tance. Third, the universal association of colors with

meanings should be taken into account. Several stud-

ies [(Dunlap et al., 1986), (Ryan, 1991), (Silver et al.,

2002)] indicate that red has the highest perceived haz-

ard level. Yellow has a medium and green has the low-

est perceived hazard level. We complied with these

ﬁndings in order to prevent a learning curve or mak-

ing the customer think about what each color repre-

sents. Additionally, the states with timeouts are coded

ParkingLotManagementforChargingStations

291

as blinking colors for an emphasis.

3.2 Example Scenarios

This section embodies two use cases where ofﬂine be-

havior of the charging station is excluded, i.e. the

charging station is connected to the central system.

Figure 2 illustrates a scenario where the customer

is registered or has a reservation. After the vehicle

parks, the charging station automatically changes to

the occupied state, the timer starts and a message is

sent to the central system. The customer requests

authorization in time with the mechanism that the

charging station supports. In this case, the charg-

ing station has an integrated RFID reader and the

customer passes his/her card in front of the RFID

reader. The system grants authorization, the charg-

ing station changes to the authorized state and the

timer starts. The customer plugs in the charging cable

in time, charging process starts and the charging sta-

tion changes to the charging state. After the battery

is charged or if the customer wishes to interrupt the

process, he/she can unplug the cable and the charging

station changes to the warning state. A timer starts

and the vehicle has to leave the parking space. But

the timer elapses and the vehicle is still parking. In

this situation, another message is sent to the central

system about the violation of terms of use.

Figure 3 illustrates a scenario where the customer

is not registered or is not the reserved customer. After

the vehicle parks, the charging station automatically

changes to the occupied state, the timer starts and a

message is sent to the central system. The customer

requests authorization in time but the system does not

grant authorization. The charging station changes to

the warning state and the timer starts. The vehicle

leaves the parking space in time, so there is no viola-

tion. A message is sent to the central system to inform

about the new status.

4 THE PROTOCOL

This section describes how our approach can be uti-

lized in an existing communication protocol between

charge points and a central system. For this purpose

we chose to demonstrate how the OCPP 1.5 can be

extended in order to support parking lot management.

It is to be implemented with Simple Object Access

Protocol (SOAP)

over HTTP for minimal overhead

and in order to maintain consistency since OCPP uses

SOAP is a speciﬁcation for encoding messages that are

exchanged in computer networks.

Central System Charging Station

1. Vehicle parks

Occupied

free

occupied

charging

violated

warning

Violated

5. Vehicle is still parking

after the timeout

authorized

2. Customer requests

authorization with RFID card:

ID is registered OR

ID == ID

reserved

3. Customer plugs in the

charging cable

4. Customer unplugs the

charging cable

Figure 2: Customer is registered or has a reservation.

Central System Charging Station

1. Vehicle parks

Occupied

free

occupied

free

warning

Free

3. Vehicle leaves

2. Customer requests

authorization with RFID card:

ID is not registered OR

ID != ID

reserved

Figure 3: Customer is not registered or is not the reserved

customer.

the same two protocols. As with OCPP, the exten-

sion uses Web Services Addressing

and contains a

Web Services Addressing is a speciﬁcation for convey-

ing addressing information of the sender and receiver in

SOAP messages.

SMARTGREENS2014-3rdInternationalConferenceonSmartGridsandGreenITSystems

292

chargeBoxId element in the SOAP message to iden-

tify the charging station.

4.1 Operations Initiated by the

Charging Station

In the following we present the necessary changes to

the protocol and how existing implementations should

be modiﬁed in order to support parking management

for the operations initiated by the charging station.

4.1.1 Protocol Changes

While automatically changing from one state to the

another, the charging station informs the central sys-

tem about the status of a parking space so that the cen-

tral system keeps track of changes. Therefore a new

operation, InformParking, should be initiated by the

charging station. We opted only to use the ﬁeld con-

nectorId (the socket identiﬁer) rather than a parking

space identiﬁer or sensor identiﬁer since the mapping

between these three entities should be unique. The

mapping from sensor identiﬁer to parking space iden-

tiﬁer to socket identiﬁer should be taken care of by

the charging point internally and is of no interest for

the central system. Additionally, using connectorId

preserves integrity since OCPP already contains the

deﬁnition of this ﬁeld.

1. InformParking

• Request:

– connectorId : int

– timestamp : dateTime

– status :

(a) Free

(b) Occupied

• Response: Empty response

4.1.2 Implementation Changes

Furthermore, the implementation of existing opera-

tions in charging station must be extended in a way

that they trigger state changes.

• Authorize: If the ID of the customer has been ac-

cepted, charging station changes from occupied to

the authorized state, or to warning otherwise.

• StartTransaction: If the transaction starts, charg-

ing station changes from authorized to charging

state.

• StopTransaction: If the transaction stops, charg-

ing station changes from charging to warning

state.

4.2 Operations Initiated by the Central

System

In the following we present the necessary changes to

the protocol and how existing implementations should

be modiﬁed in order to support parking management

for the operations initiated by the central system.

4.2.1 Protocol Changes

Operations initiated by the central system require

minimal modiﬁcation to support the new functional-

ity. The existing operation ChangeConﬁguration is

extended with following new conﬁguration keys:

1. occupiedTimeout : int

2. authorizedTimeout : int

3. warningTimeout : int

As described in Section 3, these conﬁgure the

maximum period of time in seconds a charging station

can be in occupied, authorized and warning states be-

fore it times out. Analogously, the existing operation

GetConﬁguration is extended with the same conﬁgu-

ration keys to retrieve the value of the three timeout

settings.

4.2.2 Implementation Changes

When the central system invokes the operations

RemoteStartTransaction and RemoteStopTransaction

and the charging station accepts to execute them, it

should change to charging and warning states, respec-

tively. Additionally, when the central system invokes

the following operations at the charging station, they

set the extra conditions for the free state. So the im-

plementation of the charging station must be extended

to support them.

• ChangeAvailability: If the charging station is able

to change to the requested type Inoperative, this

corresponds to the condition inoperative. If the

charging station is able to change to the requested

type Operative, this corresponds to the condition

ready.

• ReserveNow: If the charging station accepts the

reservation, this corresponds to the condition re-

served.

• CancelReservation: If the charging station is able

to cancel the reservation, this corresponds to the

condition ready and the charging station is avail-

able to charge for any one.

ParkingLotManagementforChargingStations

293

5 IMPLEMENTATION

This section describes the hardware and software im-

plementations for our solution.

5.1 Hardware

We evaluated three sensor types to monitor the charg-

ing station surroundings: A photoelectric laser sensor,

an ultrasound sensor and an inductive loop.

The photoelectric laser and ultrasound sensor

share the same operating principal. They feature a

sender, which emits a signal that is reﬂected by an

object, and a receiver, which detects the reﬂected sig-

nal. This is then interpreted as an object being nearby.

In the case of the photoelectric laser sensor this sig-

nal is a beam of light, and in the case of the ultra-

sound sensor this signal is a sound wave with a high

frequency. The photoelectric laser sensor is not cost-

efﬁcient to be installed in large scales by operators.

Furthermore, this sensor demands direct viewing an-

gle to the object, so the installation is inconvenient

since it has to be placed above the charging station.

Having the sensor outside would endanger the func-

tionality. Another drawback was that the sensor could

detect a nearby object but not the object type. Our

evaluation showed that the ultrasound sensor had the

same issues. So these two sensors are not suitable for

the charging station setup.

The best ﬁtting solution was the inductive loop,

but it is also very elaborate to install. An inductive

loop utilizes magnetic ﬁelds to sense when a large

metal object, such as a vehicle, is nearby and causes

a change in the frequency of the electrical current.

Then, this change is recognized by the detector. The

detectors are tested and evaluated, and are able to suc-

cessfully detect and inform about the occupancy sta-

tus changes. Thus, we opted to install inductive loop

vehicle detectors under the parking spaces assigned to

the charging station that is deployed at our university.

5.2 Software

We have integrated the proposed changes into our

own OCPP central system implementation SteVe

Therefore we were able to extend and modify the im-

plementation as required. SteVe is an open source

project that supports OCPP 1.2 and 1.5. It is a Web

application designed to run under Apache Tomcat and

consists of multiple Java servlets. It uses the Apache

CXF

framework for creating and receiving SOAP

abbr. for SteckdosenVerwaltung (English: socket ad-

ministration) https://github.com/RWTH-i5-IDSG/steve

http://cxf.apache.org/

messages. Moreover, CXF supports Web Services

Addressing. SteVe was tested successfully in oper-

ation.

A prototype software for the charging station has

been developed that implements our approach. It is an

implementation of the the diagram in Figure 1 as a ﬁ-

nite state machine, that supports collecting sensor in-

formation and changing LED colors according to the

state change. The communication between the charg-

ing station that is running an implementation of OCPP

and SteVe is currently a subject of work in progress.

6 CONCLUSION

We presented an integrated solution for managing

parking lots with charging stations since the existing

solutions are exclusive to either parking management

or charging station management. Our approach en-

ables the parking spaces to be occupied only by au-

thorized electric vehicles. This is achieved by tightly

monitoring and controlling user interaction with the

charging station. In addition, charging stations are

equipped with luminaries that guide customers in

search of a parking space and during the actual use

of the charging station.

Future work consists of evaluating and improving

our proposed system based on user-centric ﬁeld tests.

The architecture of our system paves way to publish

parking availability of the charging stations via the

Internet for drivers to check by mobile devices be-

fore arriving at their destination. This is due to the

fact that the central system keeps track of the park-

ing changes and the user interaction. Furthermore,

the currently-only-visual guidance system can be ex-

tended by installing speakers in the charging station in

order to provide aural assistance during the use of the

charging station. This would improve the perception

of state changes, timeouts and the run of the timer.

ACKNOWLEDGEMENTS

This work was supported by the German Federal

Ministry of Economics and Technology

(Grant 01ME12052 econnect Germany).

Bundesministerium für Wirtschaft und Technologie

(BMBF) http://www.bmwi.de/

SMARTGREENS2014-3rdInternationalConferenceonSmartGridsandGreenITSystems

294

REFERENCES

Dunlap, G. L., Granda, R. E., and Kustas, M. S. (1986). Ob-

server Perceptions of Implied Hazard: Safety Signal

Word and Color Words. Technical Report T00.3428,

IBM.

Hanabusa, H. and Horiguchi, R. (2011). A study of the an-

alytical method for the location planning of charging

stations for electric vehicles. In Proceedings of the

15th International Conference on Knowledge-based

and Intelligent Information and Engineering Systems

- Volume Part III, KES ’11, pages 596–605, Berlin,

Heidelberg. Springer-Verlag.

Hess, A., Malandrino, F., Reinhardt, M. B., Casetti, C.,

Hummel, K. A., and Barceló-Ordinas, J. M. (2012).

Optimal deployment of charging stations for electric

vehicular networks. In Proceedings of the First Work-

shop on Urban Networking, UrbaNe ’12, pages 1–6,

New York, NY, USA. ACM.

Litman, T. (2013). Parking Management: Comprehensive

Implementation Guide. Victoria Transport Policy In-

stitute.

Mehar, S. and Senouci, S. (2013). An optimization loca-

tion scheme for electric charging stations. In Smart

Communications in Network Technologies Confer-

ence, SaCoNeT ’13, pages 1–5.

OCPP Steering Group (2012). Open Charge Point Protocol

1.5. http://www.e-laad.nl/.

Ruzmetov, A., Nait-Sidi-Moh, A., Bakhouya, M., and

Gaber, J. (2013). Towards an optimal assignment and

scheduling for charging electric vehicles. In Renew-

able and Sustainable Energy Conference, IRSEC ’13,

pages 537–541.

Ryan, J. (1991). Design Of Warning Labels And Instruc-

tions. Van Nostrand Reinhold Company.

Silver, N. C., Drake, K. L., Niaghi, Z. B., Brim, A. C.,

and Pedraza, O. (2002). The Effects of Product, Sig-

nal Word, and Color on Warning labels: Differences

in Perceived Hazard. In Proceedings of the Human

Factors and Ergonomics Society Annual Meeting, vol-

ume 46, pages 735–739.

Timpner, J. and Wolf, L. (2012). Efﬁcient charging station

scheduling for an autonomous parking and charging

system. In Proceedings of the Ninth ACM Interna-

tional Workshop on Vehicular Internetworking, Sys-

tems, and Applications, VANET ’12, pages 145–148,

New York, NY, USA. ACM.

U.S. DOT (2011). Advanced Parking Man-

agement Systems: A Cross-cutting Study.

http://ntl.bts.gov/lib/jpodocs/repts_te/

14318_ﬁles/14318.pdf. Online; accessed 2014/02/10.

ParkingLotManagementforChargingStations

295