EcoLogic: IoT Platform for Control of Carbon Emissions

Tsvetan Tsokov

and Dessislava Petrova-Antonova

Department of Information Technologies, Sofia University, Sofia, Bulgaria

Department of Software Engineering, Sofia University, Sofia, Bulgaria

Keywords: Clustering, Internet of Things, Reduction of Carbon Emissions, Sensor Data Processing.

Abstract: Today, the sensors and the Internet of Things (IoT) presence naturally in the people’s lives. Billions of

interactive devices exchange information about variety of objects in the physical world. The IoT

technologies affect the business processes of all major industries such as transportation, manufacturing,

healthcare, agriculture, etc. Despite the fact that the IoT has a positive impact to both people and industry, it

also provides benefits for the environment. The IoT is recognized as a powerful tool in the fight against

climate change. More specially, it has a significant potential in saving carbon emissions. Taking into

account the promising areas of IoT application, this paper proposes a solution for real-time monitoring of

vehicles and detection of rising levels of carbon emissions, called EcoLogic. The EcoLogic consists of

hardware module that collects sensor data related to vehicles’ carbon emissions and cloud based

applications for data processing, analysis and visualisation. Its primary purpose is to control the carbon

emissions through smart notifications and vehicle’s power limitations.

1 INTRODUCTION

Today, the Internet of Things (IoT) is incorporated

in people’s lives providing an ecosystem in which

applications and services are driven by data

collected from devices that interact with the physical

world. The IoT paradigm exists in everyday physical

objects responding to human’s motion, presence,

commands and physiological behaviour. It plays a

fundamental role in economic and social

development. The combination of network

connectivity, sensors, devices and people enable a

new way of conversation between persons and

machines as well as between software and hardware

systems. The growth of sophisticated data analysis

techniques inspired by the artificial intelligence and

machine learning allows devices to anticipate, react,

respond and enhance the physical world. Advanced

applications are developed to collect and process

large amounts of data generated by the IoT devices

in all economic sectors such as transportation,

agriculture, health and education. According to the

IoT forecast of the International Data Corporation 30

billion connected devices are expected in the market

by 2020 (MacGillivray, 2016). The economic value

of IoT is evaluated around 1.46 trillion (Turner,

2016). For the same time period Gartner expects

20.8 billion connected things and 3 trillion IoT

endpoint spending (Gartner, 2015).

One of the sectors that is most affected by the

IoT is automotive industry. IoT technologies enable

production of highly automated and connected

vehicles that will change the global automotive

market. Recently, tens of millions of cars are said to

be connected to the Internet and their number is

expected to become hundreds of millions in the near

future (Automotive IT-Kongress, 2015). At the same

time, mobile communication technology is

recognized to have considerable potential to enable

carbon emissions reduction across a variety of

applications in a wide range of sectors (Stephens,

Iglesias and Plotnek, 2015). According to Global e-

Sustainability Initiative, 70% of the carbon savings

currently being made come from the use of machine-

to-machine (M2M) technologies. The greater

savings comes from buildings (29%) and

transportation (28%). The survey data shows that

68% of smartphone users are willing to adopt

behaviours that could result in even more substantial

future reductions to personal carbon emissions. IoT

is pointed as a key lever to reduce the carbon

emissions in a statistic of A.T. Kearney (A.T.

Kearney, 2015). In particular, car sharing,

automotive telematics and smart home are the most

178

Tsokov, T. and Petrova-Antonova, D.

EcoLogic: IoT Platform for Control of Carbon Emissions.

DOI: 10.5220/0006462201780185

In Proceedings of the 12th International Conference on Software Technologies (ICSOFT 2017), pages 178-185

ISBN: 978-989-758-262-2

promising cases.

The current software solutions for tracking and

monitoring vehicles give evidence for the efforts of

using IoT technologies in automotive industry.

Geotab provides a service for monitoring and

analysis of vehicles using integrated hardware

module that sends data to private cloud platform.

The hardware module is directly connected to the

onboard diagnostic system of the vehicle and

collects data about fuel consumption, travelled

distance and other parameters (GO7, 2017). The

analysis provided by the cloud platform allows

identification of vehicles with suboptimal fuel

consumption (MyGeotab, 2017). Unfortunately,

Geotab solution does not provide control over

vehicle’s parameters and detect anomalies related to

increasing rate of carbon emissions. The data logger

of Madgetech provides functionality for regular

monitoring of carbon dioxide levels (Data Loggers).

It measures the carbon emissions in exhaust system

of vehicles and sends data to private cloud platform

through a wireless network. The measured data is

visualized by mobile application, but further analysis

are not supported. In addition, functionality for

control of the carbon emissions is not provided. The

CanTrack solution provides a system for real-time

GPS tracking of vehicles (CanTrack GPS). Its

Driver Behaviour Module support driver profiling

based on 5 key driving elements including driving

style, speeding and idling. The drivers are assisted to

avoid traffic delays, blocked roads and accidents

through real-time and directional traffic information.

After a collision has been detected system alerts are

generated in order to provide accurate location

information to emergency services if required. A

drawback of the CanTrack solution is that it works

only with GPS data and does not takes into account

the vehicle’s parameters related to carbon emissions.

Inspired by the low-carbon economy roadmap of

European commission and the grate opportunity

provided by the IoT technologies for reducing the

carbon emissions, this paper proposes a solution for

real-time monitoring of vehicles and detection of

rising levels of carbon emissions, called EcoLogic.

The proposed solution includes hardware module,

which collects sensor data related to vehicle’s

carbon emissions such as air pressure, air

temperature and fuel mixture and sends it to a cloud-

based application for further analysis. The results

from the analysis are used to control the carbon

emissions through smart notifications and vehicle’s

power limitations.

The rest of the paper is organized as follows.

Section 2 presents the architecture of EcoLogic,

while Section 3 describes its components. Section 4

shows a case study that validates the feasibility of

the proposed solution. Finally, section 5 concludes

the paper and gives directions for future work.

2 ECOLOGIC ARCHITECTURE

The EcoLogic is composed of hardware modules,

which are installed on vehicles and applications

providing services, which are deployed on a cloud

platform. Its architecture is shown in Figure 1.

Figure 1: EcoLogic general architecture.

EcoLogic: IoT Platform for Control of Carbon Emissions

179

The hardware module measures several physical

parameters by sensors or extracts them from the

onboard diagnostic system of the vehicle. The data is

sent to the cloud platform. The measured physical

parameters are:

 Air/fuel ratio, which is measured by lambda

sonde sensor, which is located into the exhaust

system of the vehicle.

 Absolute pressure of the air that is consumed

by the engine.

 Temperature of the air that is consumed by

the engine.

The cloud applications are implemented as

microservices, which are designed in a platform

independent way in order to have the possibility for

deployment on different cloud platforms. The cloud

applications are communicating with a relational

database, which is provided by backing service from

the cloud platform. They process the incoming data,

store it into the database and analyse it. The

hardware modules communicate with the cloud

platform with wireless network via HTTPS or

MQTT protocols. The following physical parameters

are calculated on the base of the incoming sensor

data:

 Mass of the consumed air by the engine;

 Mass of the consumed fuel by the engine;

 Mass of the carbon dioxide emissions,

exposed into the atmosphere.

All measured and calculated physical parameters

are stored in the database. A cloud-based Analytics

application performs an anomaly detection on the

streamed data by searching for vehicles that have not

optimal amount of carbon dioxide emissions or

system failures. The anomaly detection process is

based on clustering analysis. When some vehicle is

detected by the system as an anomaly, with not

optimal amount of emissions, the hardware module

is notified automatically by the cloud platform and

hardware actuator is activated to reduce the amount

of emissions. In this way the system monitors and

controls the amount of carbon dioxide emissions in

the atmosphere in real time. The hardware modules

are equipped with three actuators:

 Liquid crystal display (LCD), which visualize

the measured and calculated physical

parameters to the driver.

 Light-emitting diode (LED), which indicates

to the driver that the amount of carbon dioxide

emissions is not optimal or there is a system

failure (not optimal parameters).

 Actuator, which controls the amount of

injected fuel in the engine and regulates the

amount of emissions.

Currently, the EcoLogic has only the display and

LED actuator. The purpose of the LED actuator is to

notify the driver to manually reduce the speed and

change the driving behaviour, which leads to

reduction of the amount of emissions.

The cloud platform provides web user interface,

which is a set of HTML5, JavaScript and CSS

resources. The web user interface is publicly

available and accessible by clients via HTTPS

protocol.

The user management of the system is composed

of two roles: driver and operator. The process flow

of the system is the following:

 Driver buys a hardware module from a dealer.

 The driver installs the hardware module into

vehicle.

 The driver registers the vehicle with the

hardware module and sensors in the system.

All components have unique identification

numbers.

 Drivers are authorized to monitor, analyze and

control their own registered vehicles.

 Operators are authorized to monitor, analyze

and control all registered vehicles by regions.

 Each driver gets score points proportional to

the amount of carbon dioxide emissions

exposed in the atmosphere by their vehicles.

Drivers can participate in greenhouse gas

trading and decrease pollution taxes with their

score points.

3 ECOLOGIC COMPONENTS

This section outlines the main components of

EcoLogic. First, the components of the hardware

module are presented. After, the cloud applications

are described.

3.1 Hardware Module

The hardware module is composed of two embedded

systems: Arduino Uno and Raspberry Pi B+. The

Arduino embedded system handles the low-level

hardware sensors and actuators in the vehicles, while

the Raspberry Pi embedded system works on higher

level and communicates with the cloud platform.

3.1.1 Arduino Embedded System

The Arduino Uno embedded system provides a

functionality to measure physical parameters,

visualize the measured parameters on 4x16 liquid

crystal display, control of actuator (light emitting

diode) and communication with Raspberry Pi

ICSOFT 2017 - 12th International Conference on Software Technologies

180

embedded system. The physical parameters are

measured by sensors or extracted from the onboard

diagnostic system (OBD2), which is provided by the

electronic control module of the vehicle. If the

vehicle provides the necessary parameters in the

onboard diagnostic interface, no additional sensors

will be installed. If the vehicle does not provide the

necessary parameters in the onboard diagnostic

interface, additional sensors, which measure these

parameters, will be installed. In this way the

hardware module is platform independent and can be

installed on different vehicles. The amount of carbon

dioxide emissions is calculated from the measured

physical parameters as follows.

The law of ideal gas (Clapeyron, 1834) is presented

with Equation 1.

     



,

(1)

where P is the absolute pressure of gas [Pa], V is the

volume of gas [m

], n is the amount of substance of

gas [mol], m is the mass of gas [kg], R

specific

is a

specific gas constant for dry air (287.058 J kg

−1

)

and T is the temperature of gas [K].

The mass of the consumed air by the engine is

calculated by the ideal gas law as shown on

Equation 2.















(2)

The air fuel ratio (AFR) is calculated according to

Equation 3:



 









(3)

The mass of the consumed fuel by the engine is

calculated by the measured air/fuel ratio (AFR) as

follows:





























(4)

The relation between mass of carbon dioxide

emissions and mass of consumed unleaded petrol

fuel is given on Equation 5 (Carbonfund, 2017).







1.73



(5)

The final equation of the mass of the carbon dioxide

emissions exposed into the atmosphere is calculated

according to Equation 6.







1.73













(6)

The application, which is deployed on the Arduino

embedded system, is implemented using C++

programming language and consumes the API

provided by the Wiring library, which is part of the

Arduino platform. The Wiring library communicates

with the appropriate microcontroller via drivers.

Currently, the Arduino Uno is used, which has

Microchip ATmega328 microcontroller with RISC

architecture.

3.1.2 Raspberry Pi Embedded System

The Raspberry Pi B+ embedded system is a proxy

between the Arduino embedded system and the

cloud platform. Its architecture is shown in Figure 2.

Figure 2: Raspberry Pi Architecture.

The Raspberry Pi B+ embedded system

communicates with the Arduino embedded system

via serial communication (UART – Universal

Asynchronous Receiver/Transmitter) over a custom

protocol. It is connected to the cloud platform via

802.11n wireless network. The Raspberry Pi B+

embedded system consumes the measured physical

parameters by the Arduino embedded system, stores

the last data in a local storage for further processing

to the cloud platform. It sends the data to Adapter

application in the cloud platform via HTTPS or

MQTT protocol and receives response that contain

information about the state of the vehicle, including

the amount of the carbon dioxide emissions. The

state can be optimal (eco) or not optimal (not eco). If

the emissions are not optimal, the Raspberry Pi

module notifies the Arduino embedded system to

EcoLogic: IoT Platform for Control of Carbon Emissions

181

activate the hardware actuator in order to reduce the

amount of emissions. The Raspberry Pi embedded

system is composed of System on a Chip (SoC) with

ARM architecture and Linux based operation system

– Raspbian. VehicleAgent application is deployed on

it. The VehicleAgent application is implemented

using Java programming language and runs on Java

Virtual Machine.

The VehicleAgent application uses the Pi4J and

WiringPi libraries, which are used for implementing

the hardware serial communication between the

Raspberry Pi and Arduino embedded systems. The

wireless communication with the cloud platform is

provided by 802.11n WiFi adapter connected to one

of the USB ports of the Raspberry Pi. It can

communicate to the cloud platform via several

application layer protocols: HTTPS or MQTT,

depending on the supported protocol by the cloud

platform. The application has a modular architecture

and enables extension with other protocols. If the

wireless network is faulty and the communication to

the cloud platform is constrained, the application

stores the last data in the local storage. After

successful connection with the cloud platform, the

locally stored data is sent to the cloud platform. The

application consumes configuration file

(config.xml), which is located into the local file

system and contains identification strings for the

vehicle, sensors and type of the communication with

the cloud platform (HTTPS, MQTT). The

configuration file should be filled by the driver or by

operator, who registers the vehicle into the system.

3.2 Cloud Applications

The EcoLogic consists of several cloud applications,

namely Controller application, Adapters applications

and Web user interface.

3.2.1 Controller Application

The Controller application is the main cloud

application, which handles all vehicles with their

hardware modules with sensors, calculates data,

stores data into database, analyze the data and

provides HTTP REST API. It is implemented using

the Java Enterprise Edition programming language

on top of JPA and Apache CXF services framework.

The Controller application has the following

functionality:

 Represents the data model by users, vehicles,

sensors and measurements. Each user has

vehicles, each vehicle has sensors, each sensor

has measurements of physical parameters.

 Manages the lifecycle of all users, vehicles,

sensors and measurements.

 Stores the data into relational database, which

is provided as a backing service by the cloud

platform. The application is platform

independent and can work with any relational

database, which provides Java connectivity.

The database contains four tables: User,

Vehicle, Sensor and Measurement.

 Calculates the mass of the carbon dioxide

emissions exposed into the air by vehicles.

 Handles the state of each vehicle: optimal

(eco) state and not optimal (not eco) state.

 Communicates with Analytics application,

which make anomaly detection – vehicles

which have not optimal amount of carbon

dioxide emissions.

 Exposes HTTP REST API which is consumed

by the Adapter applications and web user

interface.

3.2.2 Adapter Applications

The Adapter applications are cloud applications,

which adapt the data coming from the vehicle

hardware modules to the Controller cloud

application. They are implemented using the Java

programming language. There are two types of

Adapter applications that handle different types of

network protocols: ControllerAdapterHttps and

ControllerAdapterMqtt for HTTPS and MQTT

protocols respectively. ControllerAdapterMqtt

application communicates with MQTT broker. The

MQTT broker can be: Mosquitto, HiveMQ, Mosca

or other. The MQTT broker and the Adapter

applications have public URLs and can be accessed

by the vehicle hardware modules. The cloud

platform routes the traffic to the appropriate

application depending on the application layer

protocol that is used. In case of HTTPS traffic, the

cloud platform routes the traffic to the

ControllerAdapterHttps application. In case of

MQTT traffic, it routes the traffic to the MQTT

broker. This routing capabilities of the cloud

platform are based on TCP routing. TCP routing

enables cloud platforms to support applications,

which communicate with different non-HTTP

protocols. Cloud Foundry platform is an industry

standard cloud platform and it is a typical example

for platform, which uses TCP routing (Cloud

Foundry).

The lifecycle of the MQTT traffic is the

following: when a new vehicle is created, the

Controller application registers a topic with name

ICSOFT 2017 - 12th International Conference on Software Technologies

182

vehicles/{id}/sensors/{id}/measurements and

subscribes to the ControllerAdapterMqtt application

for that topic. The appropriate hardware module also

subscribes to that topic and publish measurements

related to it. The ControllerAddapterMqtt

application receives the measurements and sends

them to the Controller application, but via the HTTP

protocol. The response from the request contains

information whether the corresponding vehicle is in

optimal state or not. Finally, the

ControllerAddapterMqtt application publish the

state to the topic vehicles/{id}/state and the

appropriate vehicle is notified. In this way the

ControllerAddapterMqtt application adapts the data

from MQTT to HTTP and vice versa.

3.2.3 Web User Interface

The web user interface is provided by static

HTML5, JavaScript and CSS resources that are

served by a web server. The web resources contain

only front-end code without back-end functionality.

Most of the cloud platforms support serving of static

web resources. The web user interface in the

EcoLogic is implemented by the open-source

JavaScript-based front-end web application

framework OpenUI5 (OpenUI5). The web user

interface makes AJAX (Asynchronous JavaScript

and XML) calls to the HTTP REST API provided by

the Controller application. It provides public access

and it is used by the drivers and operators of the

system. The web user interface and the Controller

application have different domain names, which

means that they have different origins. Most of the

cloud platforms provide way for resolving the

problem of the same-origin policy (OpenUI5), which

states that one web application can access web

resources from the same origin or only permitted

web resources from different origin.

The web user interface has model-view-controller

architecture and provides the following

functionality:

 Management of the lifecycle of the users,

vehicles, sensors and measurements: performs

create, read, update, and delete operations.

 Visualization of all measured parameters and

historic data in real-time.

 Manual control of the state of all vehicles,

which leads to control of the vehicle’s

actuator.

 Automatic control of the state of the vehicles,

based on the amount of emissions or

parameters that are not optimal.

 Static limitation of emissions for all vehicles,

which can be enabled or disabled.

 Visualization of all anomalies (outliers) –

vehicles with not optimal emissions or

parameters.

3.2.4 Analytics Application

The Analytics application makes a clustering

analysis on the stored data by two parameters:

engine capacity of the vehicles and amount of

carbon dioxide emissions. In this way it places

vehicles that have adjacent engine capacity and

emissions amount in clusters and detects the vehicles

with anomalies. The analytics application uses K-

Means algorithm for clustering analysis, where the

number of clusters (K) is the number of engine

capacities of the registered vehicles. The application

is connected to the backing service with the

relational database.

4 CASE STUDY

In order to prove the feasibility of the EcoLogic

solution, a case study with two datasets is

performed. The goal of the cases study is to validate

the ability of EcoLogic to detect anomalies in the

vehicles’ behaviour related to increased carbon

dioxide emissions. One hardware module installed

on a real vehicle is used for the experiment. The

cloud platform and services that are configured for

the case study are as follows:

 SAP Cloud platform for deployment of all

described cloud applications (SAP Cloud

platform).

 HANA database – SAP cloud platform

provides backing service with HANA

relational database, which is used for storing

the data (SAP HANA database).

 SAP cloud platform predictive service,

providing an algorithm for clustering analysis

and anomaly detection, which serves as

Analytics application (Morzaria, 2016).

4.1 Validation on a Dataset with

Known Anomalies

In order to validate the correctness of the anomaly

detection algorithm it should be tested with a test

dataset, which contains known anomalies. An

official test dataset is used for that purpose

(Mugglestone, 2014). It contains information for

customers with the following parameters: id, name,

lifespend, newspend, income and loyalty. The test

EcoLogic: IoT Platform for Control of Carbon Emissions

183

dataset contains 152 rows. The results from

application of the clustering analysis on the test

dataset are presented in Figure 3.

Figure 3: Clusters of Dataset with Known Anomalies.

The income of the customers is placed on the x-

axis. The loyalty of the customers is placed on the y-

axis. Two clusters (K1 and K2) and one anomaly

detection are obtained. The clusters define two types

of customers: customers with low income and low

loyalty and customers with high income and high

loyalty. The anomaly, marked in Figure 3 as Outlier,

corresponds to a data point, which is outlying from

the centres of the clusters K1 and K2.

4.2 Validation on a Real Dataset

The hardware module of the EcoLogic is integrated

into a real vehicle with internal combustion engine

that works on petrol and has a capacity of 1800

cubic centimetres. The collected real data for it is

extended proportionally with appropriate simulated

data in order to obtain bigger dataset. The final

dataset contains data for vehicles with different

engine capacities. The engine capacity measured in

cubic centimetres (cc) is placed on the x-axis. The

mass of the carbon dioxide emissions, measured in

milligrams (mg) is placed on the y-axis. The

obtained clusters after application of the clustering

algorithm are presented in Figure 4. The data points

corresponds to the vehicles, which have unique IDs.

Figure 4: Clusters of Real Dataset.

In the most preferable case, the clustering

algorithm should place vehicles, which have equal

engine capacity and different amount of emissions in

the same cluster. Thus, the vehicles, which have not

optimal emissions will not be placed in cluster and

should be detected as anomalies. For the current

dataset 8 clusters (K1-K8) are obtained. The

distance from the vehicle with ID 8759305 to the

nearest cluster K4 is 38 milligrams. The maximal

internal cluster distance in the cluster K4 is 18

milligrams. The distance from vehicles with IDs

6947228 and 5915180 to their nearest cluster K8 is

54 and 56 milligrams respectively. The maximal

internal cluster distance in cluster K8 is 24

milligrams. The anomalies are detected for the

vehicles with IDs 8759305, 6947228 and 5915180,

since the distance from them to their nearest clusters

is bigger than the internal cluster distance. These

vehicles don’t have optimal amount of carbon

dioxide emissions in contrast to the rest of the

vehicles, which are placed into clusters K1-K8. The

values of the anomalies are measured on cold engine

of the vehicle. The hardware modules successfully

notify the actuators for the detected anomalies.

5 CONCLUSIONS

The paper presents an IoT platform, called

EcoLogic, for real-time monitoring and control of

carbon dioxide emissions of vehicles with internal

combustion engines.

It has the following benefits:

 High scalability, resilience and possibility to

work with big amounts of data due to cloud

computing model used.

 Platform independence – possibility to work

with different vehicles and cloud platforms.

The hardware modules can work with variety

sensors or extract data from the onboard

diagnostic system of the vehicles. The

implemented cloud applications are

microservices, which can be deployed on

different cloud platforms.

 Fully completed solution for monitoring and

control of vehicles’ carbon dioxide emissions,

which is ready for production usage to solve a

global problem such as reduction of the

carbon dioxide emissions in the atmosphere.

The following directions for further improvements

are identified during the validation on the real

dataset:

 The measured parameters on cold engine are

not consistent with the measured parameters

ICSOFT 2017 - 12th International Conference on Software Technologies

184

on hot engine (with normal working

temperature). The detected anomalies are

obtained based on measurements on a cold

engine. The EcoLogic solution could be

optimized to split the data in two subsets: data,

which is taken on cold engine and data, which

is taken on normally working engine.

 The data, which is taken from cold engine

could serve as a training dataset, which

defines not optimal amount of carbon dioxide

emissions.

 Integration of new application protocols such

as CoAP, DDS and AMQP that can be used

by default.

 Implementation of analytics functionality for

prediction of potential failures in vehicles,

based on the current and historical data.

 Intergradation of EcoLogic with third party

systems and services such as emissions

trading systems, vehicle tax institutions and

smart cities systems. For example, the drivers

of the more ecological vehicles could get

bigger tax discounts. The traffic lights in

smart cities could be controlled depending on

the amount of carbon dioxide emissions

detected in a region. Adaptation mechanisms

for service selection based on criteria like

geographic location, price, load balancing, etc.

will be considered.

ACKNOWLEDGEMENTS

The authors acknowledge the financial support by

the National Scientific Fund, Bulgarian Ministry of

Education and Science within the project no. DN

02/11/2016 and project no. DFNI I02-2/2014, and by

the Scientific Fund of Sofia University within

project no. 80-10-192/24.04.2017.

REFERENCES

MacGillivray, C., 2016. Worldwide Internet of Things

Forecast Update. 2015-2019, International Data

Corporation (IDC).

Turner, V., 2016. The Internet of Things: Getting Ready to

Embrace Its Impact on the Digital Economy.

International Data Corporation (IDC), 2 March 2016,

Directions 2016: San Jose, CA.

Clapeyron, E. (1834). Mémoire sur la puissance motrice

de la chaleur. Journal de l'École Polytechnique (in

French).

Gartner, 2015. Gartner Says 6.4 Billion Connected

‘Things’ Will Be in Use in 2016, Up 30 Percent From

2015. Gartner website, 10 November 2015, Available

at: http://www.gartner.com/newsroom/id/3165317

(Accessed 8 Apr. 2017).

Automotive IT-Kongress 4.0, 2015, Available at:

http://www.t-systems.de/news-media/automotiveit-

kongress-industrie-4-0-veraendert-

automobilindustrie/1339486 [Accessed 4 Apr. 2017]

Stephens A., Iglesias M., Plotnek J., 2015. GeSI Mobile

Carbon Impact, Available at:

http://gesi.org/files/Reports/GeSI%20mobile%20carbo

n%20impact.pdf (Accessed 2 Apr. 2017).

A. T. Kearney, 2015. Internet of Things, a key lever to

reduce CO2 emissions, Available at:

http://www.atkearney.fr/documents/877508/879237/2

0151113_IoT+Impact+on+energy_Europe+EN.pdf/67

57111f-21da-49ee-82fd-915ff42dc26d (Accessed 6

Apr. 2017).

GO7. (2017). UK: Geotab.

MyGeotab. (2017). UK: Geotab.

Data Loggers. USA: Madgetech.

CanTrack GPS. UK: CanTrack Global Ltd.

Mugglestone Ph., (2014). SAP HANA Academy. Available

at: https://github.com/saphanaacademy/PAL/tree/

master/Source%20Data/PAL (Accessed 6 Feb. 2017).

OpenUI5. (2017). SAP SE.

Cloud Foundry. Enabling TCP Routing. Available at:

https://docs.cloudfoundry.org/adminguide/enabling-

tcp-routing.html.

SAP Cloud platform. (2017). SAP SE.

SAP HANA database. (2017). SAP SE.

Morzaria, A. (2016). Announcing “SAP HANA Cloud

Platform predictive services”. [Blog] SAP Blogs,

Available at: https://blogs.sap.com/2016/02/18/

announcing-sap-hana-cloud-platform-predictive-

services/ (Accessed 18 Feb. 2017).

Carbonfund. (2017). How we calculate. Available at:

http://www.carbonfund.org/how-we-calculate

(Accessed 12 Jan. 2017).

EcoLogic: IoT Platform for Control of Carbon Emissions

185