myBee: An Information System for Precision Beekeeping

Luis Gustavo Araujo Rodriguez

, Jo

ao Almeida de Jeus

, Vanderson Martins do Ros

ario

Anderson Faustino da Silva

, Lucimar Pontara Peres

, Heberte Fernandes de Moraes

and Claudio Luis de Amorim

Department of Informatics, State University of Maring

a, Colombo Avenue, 5790 - Block C56, Maring

a, PR, Brazil

Department of Zootechny, State University of Maring

a, Colombo Avenue, 5790 - Block J45, Maring

a, PR, Brazil

Parallel Computing and Mobile Systems Laboratory, Federal University of Rio de Janeiro,

City University - Block I - Room I-232, Rio de Janeiro, RJ, Brazil

Keywords:

Mobile Computing, Information Systems, Precision Beekeeping, Computing Platform, Wireless Computing.

Abstract:

Over the last few years, beekeepers have become aware of the necessity of integrating Information Technology

into Apiculture. As a result, Precision Beekeeping emerged, which is an area that applies the said technology to

monitor bees and, consequently, the state of the colony. However, the development of these platforms is com-

plex due to the requirement of them adapting to various heterogeneous environments and constantly-updated

technologies. Thus, the objective of this paper is to propose and develop a Precision Beekeeping Information

System, called myBee, whose infrastructure is ﬂexible, secure, fault tolerant, and efﬁcient in decision-making.

A real case study shows that myBee is an efﬁcient information system that supports beekeepers in maintaining

their apiaries.

1 INTRODUCTION

A healthy ecosystem is indispensable for human life,

animals and natural resources. Honey Bees are nat-

ural sensors of ecosystems and one of the most im-

portant insects. They possess the ability to pollinate

(Magno et al., 2015; Murphy et al., 2015), conse-

quently providing nutrients for humans and animals,

as well as aiding in ecosystem health. Thus, it is in-

dispensable to preserve them, as they play a major

contribution to ecosystems, as well as to the global

economy (Zacepins et al., 2016).

Recently, beekeepers have perceived the necessity

of adopting Information Technologies (IT) into the

agricultural ﬁeld (Kviesis and Zacepins, 2016). This

occurs not only for the important role of honey bees in

crop production, but also due to the fact that bee pop-

ulation has been decreasing over the last few years

(Kviesis et al., 2015; Murphy et al., 2015; Zacepins

et al., 2016). Therefore, monitoring these insects has

become a crucial activity (Magno et al., 2015; Za-

cepins et al., 2016).

In this context, Precision Beekeeping (PB)

emerged, which is a subdivision of Precision Agricul-

ture that applies IT in order to determine the state of

the bee colony and improve its preservation (Zacepins

et al., 2015).

Since bees are important worldwide (Murphy

et al., 2015), several Information Systems (IS) have

been developed (Zacepins et al., 2015; Kviesis et al.,

2015) and applied to PB. However, it is worth men-

tioning that although different platforms have been

implemented, PB is still in development stages (Za-

cepins et al., 2015; Magno et al., 2015; Kviesis and

Zacepins, 2016; Zacepins et al., 2016). In addi-

tion, the majority of Precision Beekeeping Informa-

tion Systems (PBIS), proposed in the literature, do not

include the distinguishing characteristics of a well-

developed IS.

A well-developed IS has the following character-

istics:

• simpliﬁes the development and maintenance pro-

cess;

• provides reusability of modules and subsystems;

• has modules and subsystems that are plug-and-

play;

• provides compatibility for different devices;

• provides compatibility for stationary and mobile

systems;

Rodriguez, L., Jeus, J., Rosár io, V., Silva, A., Peres, L., Moraes, H. and Amorim, C.

myBee: An Information System for Precision Beekeeping.

DOI: 10.5220/0006285205770587

In Proceedings of the 19th International Conference on Enterpr ise Information Systems (ICEIS 2017) - Volume 2, pages 577-587

ISBN: 978-989-758-248-6

577

• provides security; as well as,

• provides quality of service.

The objective of this paper is to describe the well-

developed Precision Beekeeping Information System

(PBIS), called myBee, whose objective is to support

beekeepers in maintaining their apiaries. A real case

study shows that myBee is an efﬁcient PBIS.

The rest of the paper is organized as follows. Sec-

tion 2 presents related works in regards to Precision

Beekeeping and Information Systems. Section 3 de-

scribes the PBIS, myBee. Section 4 presents myBee

in a real-world environment. Finally, Section 5 dis-

cusses the conclusions and future works.

2 RELATED WORKS

The following related works are divided into two cat-

egories: Precision Beekeeping Systems and Informa-

tion Systems.

2.1 Precision Beekeeping Systems

Armands Kviesis et al. (2015) proposed a PB sys-

tem platform that utilized a SHT15 sensor to measure

temperature and humidity. A total of eight beehives

were placed outdoors, each with measurement nodes

in closed boxes and protected with waterproof mate-

rial. All temperature and humidity data were stored

in a SQL database, which can be visualized in a web

application.

Gatis Riders et al. (2015) proposed a decision-

support module to better-comprehend the collected

data and, consequently, execute reliable actions. The

sensor utilized in the experiment was DS18S20,

which measures temperature. The sensors were con-

nected to a Raspberry Pi and data was sent to a server

and stored in a MySQL database. Two systems were

developed, the ﬁrst monitored ten honey bee colonies

inside a wintering building, while the second moni-

tored ten honey bee colonies outside. The data can be

visualized in either a web or desktop application, pro-

viding the option to view the maximum, minimum,

median and average temperatures, by day, for all in-

stalled sensors. Both systems have a decision-support

and analysis module.

Marco Giammarini et al. (2015) developed a mon-

itoring system to collect temperature and humidity

data of two beehives in the summer. One beehive was

placed in a wooden box, while the other in a plastic

box. A GSM modem was implemented for sharing

and downloading data; remote monitoring; and soft-

ware debugging.

Fiona Edwards Murphy et al. (2015) imple-

mented a system, utilizing a Wireless Sensor Network

(WSN), to monitor a bee colony. The platform col-

lected data such as temperature, carbon and nitrogen

dioxide, humidity, pollutants and battery percentage

of the device being utilized. The researchers applied

WSN due to being a non-intrusive technology, thus,

allowing more accurate data collection. The informa-

tion can be accessed through a web interface or mo-

bile device.

Michele Magno et al. (2015) implemented a pro-

totype to collect images and audio within a beehive.

The platform utilizes a Libelium Waspmote and Rasp-

berry Pi in order to process and store the data. Ad-

ditionally, microphones, accelerometers, thermal and

infrared cameras were used, along with emergency

notiﬁcations to the user in case an undesirable event

occurs to the beehive. The goal was to utilize the said

equipments to collect data in a discreet manner.

2.2 Information Systems

Wilson Goudalo et al. (2016) mention how, currently,

Information Systems play a signiﬁcant role for En-

terprises. Thus, they proposed various methods to

provide simple user interfaces for managing security.

The authors applied seven principles of the ISO 9241-

11 (ISO, 1998), which are: clarity, discriminabil-

ity, brevity, consistency, detectability, readibility and

comprehensiveness. All these principles, including

security concerns, were all taken into consideration

when implementing myBee.

Delﬁna Soares et al. (2014) deﬁned interoperabil-

ity as a characteristic in which entities preserve au-

tonomy and independence. Thus, myBee’s architec-

ture and protocol can be categorized as interoperable

due to exchanging information while being indepen-

dent from one another. The authors also recalled the

importance of a social-technical perspective for infor-

mation systems, which we believe applies to myBee as

well, due to offering data reports that inﬂuence user-

decisions and system operations.

Hadi Kandjani et al. (2013) proposed a frame-

work to classify system-planning methodologies. The

authors stated that selecting a proper methodology

to develop an information system is a key factor for

its success. We believe the methodology utilized

to implement myBee was a success due to the ob-

tained observations. In addition, the authors men-

tioned that several methodologies for implementing

and planning information systems are available, thus,

myBee’s architecture is ﬂexible to be utilized in vari-

ous ﬁelds, which simpliﬁes and streamlines the devel-

opment process.

ICEIS 2017 - 19th International Conference on Enterprise Information Systems

578

Ovidiu Noran (2013) proposed enhancements,

based on interoperability, to Disaster Management In-

formation Systems, using an enterprise architecture

perspective and artifacts. The author claimed that

these types of systems are important due to allow-

ing a collaboration for environmental incidents. This

statement can also be applied to myBee, which pro-

vides crucial data for beekeepers and, therefore, col-

laborates and supports the environment. In addition,

myBee’s architecture is based on interoperability, al-

lowing communication and easy access between each

component.

Jorge Aguiar et al. (2013) proposed an improve-

ment for Decision Support Systems corresponding to

Intensive Care Units based on TAM (Technology Ac-

ceptance Model). The architecture for an ICU infor-

mation system can be divided into two subsystems:

one to collect the data and another to process and

display data. This statement can also be applied to

myBee, which is divided into two ﬂexible and efﬁcient

subsystems (stationary and mobile). As in (Aguiar

et al., 2013), we proposed improvements to an infor-

mation system, which will contribute on improving

and supporting a speciﬁc area (Precision Beekeep-

ing).

3 myBee

This paper proposes a PBIS called myBee, which is

based on two approaches described by Kviesis & Za-

cepins (2015):

1. using an interface device for each beehive; and

2. sending data to a remote computational center.

Therefore, this platform will offer a more detailed

analysis of the bee colonies, resulting in a better main-

tenance and preservation of Honey Bees.

A well-developed IS requires a ﬂexible (Rabaey

et al., 2006), scalable and efﬁcient architecture. In

fact, Zacepins & Stalidzans (2012) claim that infras-

tructures with sub-elements are needed for PB. Thus,

myBee’s architecture was developed pursuing the fol-

lowing characteristics:

1. Flexibility: the PBIS has to be based on simplic-

ity. As a result, myBee can be easily maintained

and modiﬁed to better suit the beekeeper’s needs.

2. Fault Tolerant: the PBIS has to efﬁciently han-

dle potential errors. Thus, myBee was imple-

mented with several precautionary measures in-

cluding data redundancy.

3. Security: the PBIS has to provide security mecha-

nisms to ensure that the data is not violated. Thus,

myBee uses a secure protocol.

4. Efﬁciency in Decision-making: the PBIS has to

simplify the maintenance of bee colonies. As a

result, myBee provides several reports, estimates

future conditions and warns about undesirable be-

haviours.

Figure 1 outlines myBee.

Figure 1: PB System Platform.

As indicated in Figure 1, myBee is divided into two

subsystems:

1. the Stationary System; and

2. the Mobile System.

The Stationary System consists in collecting and

storing the data. The Mobile System consists of the

software to monitor the collected data. A brief de-

scription of myBee is as follows: DHT22 sensors lo-

cated at the center of the beehives monitor the condi-

tions, collecting pieces of information, which are sent

through a wireless network to a server. Therefore, us-

ing the Mobile System beekeepers can monitor, via

a web interface on a computer or mobile device, the

conditions of the bee colonies.

3.1 The Stationary System

The Stationary System, whose objective is to provide

a well-deﬁned and developed IS, is a three-layered ar-

chitecture divided by a middleware component. The

architecture was designed to provide ﬂexibility and

simplicity in accessing the layers. Thus, the mid-

dleware component is the most important and distin-

guishing characteristic of the proposed infrastructure.

In addition, a Management/MachineLearning compo-

nent permits the user to conﬁgure or modify the op-

erations, protocols, storage and physical components.

Furthermore, it estimates future conditions of the bee

colony.

Figure 2 displays the layered-based infrastructure

for the Stationary System platform.

It is worth mentioning that the Stationary System

architecture can be utilized in a wide range of sce-

narios and is not limited to the PB branch. As stated

myBee: An Information System for Precision Beekeeping

579

Figure 2: Stationary System Architecture.

before, this is due to the middleware, which provides

scalability and ﬂexibility.

3.1.1 The Layers

Service Layer. The Service Layer focuses on the

three basic operations of an IS:

• Monitoring/Input: consists in collecting the data

and then sending it to the Processing module.

• Processing: consists in manipulating the col-

lected data. In addition, this module determines

the type of output, whether it will be stored, sent

to the base station/server, or both.

• Transmission/Output: consists in transmitting

the information, using the protocol in the Network

Layer, to the server.

Network Layer. The Network Layer offers func-

tionalities such as protocol conﬁguration, and en-

abling and disabling data transmission. myBee was

designed to offer ﬂexibility, scalability and compati-

bility. Thus, the user can add several protocols to the

PBIS, which are handled by the Management module.

Hardware Layer. The Hardware Layer offers func-

tionality to add, conﬁgure, enable and disable the

physical components, which are also controlled by the

Management module. The hardware devices consist

of microcontrollers, sensors and transmitters.

3.1.2 The Implementation

In order to implement a software architecture that is

ﬂexible and simple to maintain, a class structure was

designed and specialized to add speciﬁc functionali-

ties (components). Thus, it is possible to modify the

architecture to provide new functionality, such as us-

ing a different communication protocol, or even sup-

porting a different device. Basically, the software ar-

chitecture has the following classes:

• Monitor: provides the functionality to monitor a

certain condition of the environment.

• Process: provides the functionality to process the

monitored data.

• Transmit: provides the functionality to receive

and transmit the data.

• Protocol: provides the functionality for communi-

cation between devices.

• Microcontroller: provides the functionality to

manage a microcontroller.

• Sensor: provides the functionality to manage a

sensor.

• Transmitter: provides the functionality to manage

the transmitter/receiver.

• Middleware: provides the functionality for com-

munication between the layers.

• Management: provides the necessary functional-

ity to manage the entire architecture, as well as the

computational intelligence.

In fact, the aforementioned classes are abstract,

meaning they must be specialized to provide concrete

functionality, with the exception of the Middleware

and Management classes.

3.1.3 Some Details

myBee has a basic functionality with specialized ab-

stract classes that compose the software architecture.

This functionality allows myBee to: use low-cost de-

vices and a database system manager; monitor the

temperature and humidity of the beehives; organize

monitoring-devices in a mesh network; provide re-

ports and statistics; and anticipate future behaviour.

Temperature and humidity sensors were chosen due

to being the most appropriate way to monitor bee

colonies (Zacepins and Meitalovs, 2014).

Hardware. Currently, the Stationary System sup-

ports a Raspberry Pi, DHT22 sensor, and the micro-

controllers GPIO7 and GPIO18. This indicates that

the said system uses low-cost hardware, which is able

to: deactivate the sensor DHT22 for a certain time;

collect the temperature and humidity of the beehive;

store data; and transmit/receive data.

Database. The Stationary System supports the

database management system, MySQL. The data is

stored on a database as shown in Figure 3.

Figure 3: Database.

ICEIS 2017 - 19th International Conference on Enterprise Information Systems

580

Protocol. The monitoring-devices that compose the

Stationary System automatically organize themselves

as a mesh network.

A mesh network differs from a traditional network

because each node has the responsibility of serving as

an access point. Thus, each node behaves as a router

and, therefore, composes a single network which can

be accessed from any point.

Since mesh networks have several access points,

this allows for a simpler implementation and higher

fault-tolerance. This is because the network adapts

to movements, inclusion and even exclusion of nodes

automatically - without the need to reconﬁgure them

(Lent, 2008).

In a mesh network, the packet jumps from one

node to another until it reaches its ﬁnal destination.

Thus, a node does not need to be visible - if it is within

reach - for the data to be sent.

A technique to organize a mesh network is to

utilize an Interest Ad-hoc Network (Radnet), which

communicates through a protocol based on interests.

Radnet is a network model based on user interests

and characteristics, and permits:

• a collaboration between the network nodes. The

messages are delivered by jumping between the

nodes;

• a routing based on user characteristics; and

• a message delivery approach to users who have

the same interests.

Radnet belongs to the Publisher/Subscriber

model, which is an asynchronous model that consists

of a publisher node sending the messages to a certain

interest and all the subscribers of that respective in-

terest receiving them. Although some messages may

not be delivered due to communication errors (Dutra,

2012).

The address of network devices and users is per-

formed through the Active Preﬁx (Dutra et al., 2012),

which is divided into two ﬁelds, namely: preﬁx; and

interest.

The preﬁx, which represents user characteristics,

is used as an Internet Protocol (IP) address. How-

ever, the difference is that the Preﬁx is linked to each

application and not to the equipment. Unlike the IP,

several users can have exactly the same preﬁx. The

interest is a ﬁeld that stores and represents an interest

of the application.

The Radnet Protocol is responsible to forward the

messages between the network devices (Dutra et al.,

2010; Dutra and Amorim, 2010). A neighbor node

can return a message that is being forwarded. Thus,

the message can be received by a node that processed

it. To avoid reprocessing, every message is inserted,

at a certain time, into the hash table. Therefore, the

table is consulted whenever a message is received.

The aforementioned protocol was developed for

mobile and low-power applications. Its main features

are: energy-saving; and sending packages indirectly

to the receiver and through multiple nodes that have

common interests. A mesh network adapts to node

failures or removals. As a result, Radnet guarantees

that the network will automatically conﬁgure itself if

nodes are added or removed (Dutra et al., 2012).

It is worth highlighting that Radnet was devel-

oped with security in mind, attaining Active Preﬁx

messages by either cryptographic signatures or pass-

words (Dutra et al., 2012). Based on these advan-

tages, this protocol was chosen for our PBIS.

Computational Intelligence. myBee provides in-

formation about possible-future values of the moni-

tored elements, which is performed by the Computa-

tional Intelligence. Basically, when the said function-

ality is triggered, the Stationary System activates arti-

ﬁcial intelligence algorithms to estimate values of the

monitored data, from the existing database. When-

ever an inference is made, the data will be validated

as soon as the system obtains it. If the inference is

wrong, a warning is issued to the user.

Client/Server Side. The characteristics of the Sta-

tionary System architecture allow both clients and

servers to be created. The former monitors the en-

vironment and sends the data to the server. The latter

persists the data into a database and sends it to a re-

mote computer or storage.

3.2 The Mobile System

The Mobile System has three objectives:

1. visualize the data monitored by the Stationary

System;

2. provide data and statistics; and

3. provide notiﬁcations about undesirable be-

haviours.

The Mobile System architecture is composed of 2

layers: Service and Management. Figure 4 displays

the layered-based architecture for the Mobile System

platform.

Figure 4: Mobile Architecture.

myBee: An Information System for Precision Beekeeping

581

3.2.1 The Layers

Service Layer The Service Layer focuses on three

aspects:

• Monitoring: consists in monitoring real-time

data on a mobile device.

• Reports: provide information based on the col-

lected data.

• Decision-making: permits user decisions based

on reports.

Management Layer. The management layer is

composed of a middleware, which connects the Mo-

bile System and Stationary System.

3.2.2 The Implementation

The Mobile System consists of a Web interface to

monitor the bee colonies. Figure 5 displays the Web

System.

The Web System has the following functionality:

• displays the monitored data;

• generates graphs;

• provides reports;

• provides statistics;

• provides notiﬁcations; and

• provides future-behaviour statistics.

The system-visualization module graphically dis-

plays the collected data, allowing the user to select

the time period. The report module provides detailed

reports, including: description of each beehive, loca-

tion, monitoring time and date. Both graphs and re-

ports can be exported, with different extensions, and,

thus, used separately.

The statistics module provides numeric values of

the monitored data, which are: general mean, stan-

dard deviation, variance, minimum and maximum

value. The ﬁrst three can be viewed in reports, while

the remaining two are displayed graphically.

The said functionality can be applied to each Sta-

tionary System individually. Thus, a single Mobile

System is used to monitor all Stationary Systems in-

stalled on different beehives. It is worth mentioning

that the Web System provides ﬁve data ﬁlters, which

are applied when viewing the data, and they are:

• all data monitored thus far;

• data monitored between 12:00 AM and 6:00 AM

(0h - 6h);

• data monitored between 6:00 AM and 12:00 PM

(6h - 12h);

• data monitored between 12:00 PM and 6:00 PM

(12h - 18h);

• data monitored between 6:00 PM and 12:00 AM

(18h - 24h);

The notiﬁcation module provides warnings to the

user, indicating that the conditions of the bee colony

are undesirable. Thus, the user can make appropriate

decisions depending on the issued notiﬁcations.

Finally, the Web System provides the user with

information estimated by the Computational Intelli-

gence module that exists in the Stationary System.

4 CASE STUDY

Beekeeping is a sustainable activity that generates

positive impacts on social, economic and environ-

mental areas. This activity provides: income to bee-

keepers though the commercialization of their prod-

ucts; beneﬁts to the environment through pollination,

which favors the balance and maintenance of biodi-

versity (Camargo et al., 2002).

Controlling temperature and humidity is essential

because biological processes can be modiﬁed and/or

altered by high variations. Thus, it is important to

implement technologies in order to maintain adequate

beehive conditions.

Therefore, the objective of the case study is to

monitor the internal temperature and humidity of the

beehive corresponding to the Apis Mellifera species.

4.1 Experiment Area

myBee is currently used in the Experimental Farm of

Iguatemi (EFI), viewed in Figure 6. The EFI is lo-

cated at a latitude of 23

◦

25’ S; 51

◦

57’ O, an altitude

of 550 meters and area of 170 hectares. This location

provides a suitable environment to develop projects

on agriculture and animal husbandry.

The apiary is composed of 10 beehives, which are

arranged in two types of boxes:

• Styrofoam and

• Wood.

Different materials are used in order to evaluate

the conditions of the beehives, each with distinctive

treatments.

The boxes were certiﬁed by the Forest Steward-

ship Council (FSC) and arranged in contrast to one

another. These boxes were directly exposed to the

weathering of the climate, reducing interference of

ICEIS 2017 - 19th International Conference on Enterprise Information Systems

582

(a) Realtime

(b) History

Figure 5: The Web Interface.

myBee: An Information System for Precision Beekeeping

583

Figure 6: Experimental Farm of Iguatemi, Source: Google

Earth.

non-climatological factors in the experiment. The ex-

periment area is surrounded by an eucalyptus plan-

tation. The beehives were homogenized and stan-

dardized, being considered for: two combs, three

with eggs and larvae, three with reserved food and

two with alveolate wax (comple plaque), totalling 10

combs for each beehive. It was necessary to feed the

bee colonies with water and sugar syrup in a 1:1 ratio

during the standardization of the swarms (Camargo

et al., 2002).

The placement of the Wooden and Styrofoam

boxes can be visualized in Figure 7.

Figure 7: Apiary.

The red-dotted line represents data transmission.

This means that each collected information is sent to

every node, through Radnet, for backup purposes. In

other words, this means that each Stationary System

was implemented as a server. This data-redundancy

guarantees that the data will be stored regardless of

equipment failure.

4.2 Observations

This section presents data collected over a period of

10 days. The data was analyzed using the method

of Least Square with the Statistical Analysis System.

The adjusted averages were compared with Tukey’s

Test (P≤0.05).

Controlling the internal temperature is important

for several reasons. Boyle-Makowski (1987), for ex-

ample, stated that the colony’s increased internal tem-

perature has signiﬁcant inﬂuence on bees and their

forage throughout the day. It has been shown that,

while worker bees can survive temperatures above

◦

C (Coelho, 1991), temperatures above 36

◦

C dur-

ing an extended period of time can cause death or ab-

normal development of the Apis mellifera offspring

(Winston, 1991).

The analysis of variance in Table 1 showed that

there was a signiﬁcant difference for each type of box,

the materials and periods corresponding to both the

temperature and relative humidity (P>0.05). There

was a small contrast (P>0.05) between the relative

humidity, whose average was 60% for the entire data-

collection period.

Table 1: Average temperature and relative air humidity (*:

signiﬁcant (P<0.05); NS: not signiﬁcant (P>0.05)).

Source of variation Degrees of T

◦

Freedom

Collections 1 0.45* 19.39

Days (Collection) 10 0.08

48.71*

Period 3 0.26* 176.98*

Material 1 0.29* 855.49*

Box (Material) 7 0.81* 615.77*

Residue 374 0.05 21.15

- 0.30 0.43

Coefﬁcient of Variation (CV) - 0.65 7.67

According to Southwick (1985) & Stabentheiner

et al. (2003), endothermic heat production and site

isolation allow bees to regulate the temperature of the

breeding chamber within the range of 32-36uC.

The values described above corroborate with those

obtained in Figure 8. Both materials did not inter-

fere in the temperature homeostasis of the swarm, al-

though a signiﬁcant difference was acquired.

1 2 3 4 5 6 7 8 9 10

Day

Temperature (C)

34.0

34.5

35.0

35.5

36.0

Wood Styrofoam

Figure 8: Temperature.

ICEIS 2017 - 19th International Conference on Enterprise Information Systems

584

The humidity, as shown in Figure 9 was signiﬁ-

cant, obtaining a higher stability for swarms placed in

a styrofoam box.

1 2 3 4 5 6 7 8 9 10

Day

Humidity (%)

Wood Styrofoam

Figure 9: Humidity.

According to Seeley (2006), controlling the tem-

perature of a beehive can be seen as one of the greatest

innovations.

Honey bees, between the beginning of autumn and

the end of winter (annual period of bee development),

maintain the temperature in the middle of the beehive

between 33 and 36

◦

C, with an average of approxi-

mately 34, 5

◦

C, and usually ranging at least than 1

◦

per day (Hess, 1926; Himmer, 1927).

The values obtained in the four periods of 6

hours/day, as shown in Figure 10 and 11, were within

the values of homeostasis and showed that a higher

temperature is expected during the night.

In terms of bees, thermoregulation is possible

within certain limits. Body-warming occurs by ab-

sorbing heat from the environment. Thus, bees can

raise their internal temperature above the ambient

temperature. In addition, musculature contractions

also contribute to heat-generation inside the body of

the said insect. This thermoregulatory potential seems

to emerge early in the worker bee’s life, as they gen-

erate heat only after a few days of being born (Esch,

1976; Heinrich, 1975).

The analysis of variance showed that there was a

signiﬁcant different between the boxes, materials and

periods for both air temperature and relative humidity

(P>0.05).

There was a small contrast (P>0.05) between the

relative humidity, whose average was 60% for the en-

tire data-collection period.

4.3 Validation

An infrared camera-system was utilized to analyze the

internal temperature of the beehive, as shown in Fig-

ure 12.

The use of such a strategy has two problems:

1 2 3 4 5 6 7 8 9 10

Day

Temperature (C)

34.0

34.5

35.0

35.5

36.0

36.5

0−6h 6−12h 12−18h 18−24h

(a) Temperature

1 2 3 4 5 6 7 8 9 10

Day

Humidity (%)

0−6h 6−12h 12−18h 18−24h

(b) Humidity

Figure 10: Wood.

1 2 3 4 5 6 7 8 9 10

Day

Temperature (C)

34.0

34.5

35.0

35.5

36.0

0−6h 6−12h 12−18h 18−24h

(a) Temperature

1 2 3 4 5 6 7 8 9 10

Day

Humidity (%)

0−6h 6−12h 12−18h 18−24h

(b) Humidity

Figure 11: Styrofoam.

1. the use of infrared images is susceptible to a

known error of 2%; and

2. the user plans to handle each beehive to identify

its condition.

Comparing the data monitored by myBee with the

myBee: An Information System for Precision Beekeeping

585

Figure 12: Infrared image of a part of the beehive.

data collected by the infrared camera, it is demon-

strated that myBee has the following advantages:

• the data monitored by myBee are statistically the

same as those obtained by the infrared camera;

• the collected data is not susceptible to an error

percentage;

• there is no need to handle each beehive to identify

its condition;

• there is no need to process the data to obtain re-

ports and statistics;

• an unfavorable condition in the beehive is known

in real time.

Using myBee in a real-world environment demon-

strates its ability to monitor the conditions of the bee

colony in order to perceive the alterations and to ana-

lyze the effects of environmental variables (tempera-

ture, humidity) on different boxes.

5 CONCLUSIONS

Currently, beekeeping provides an income of billions

of dollars. Thus, efﬁciently maintaining bee colonies

has become a goal for beekeepers. Temperature and

humidity are among the environmental variables that

affect the said area. Several techniques and technolo-

gies are used to capture, analyze and monitor bee

colonies.

In order to monitor the said variables efﬁciently

and in real-time, this paper describes myBee, which

is a well-developed PBIS that supports beekeepers in

maintaining their apiaries.

The evaluation of myBee in a real-world environ-

ment proves its efﬁciency in monitoring bee colonies,

taking into consideration climatic alterations and the

effects of different types of boxes.

A future challenge is to add sensors to the system

in order to monitor and analyze sound and weight of

the bee colony.

ACKNOWLEDGEMENTS

We offer our sincere gratitude to CNPq and UFRJ for

their support. This article would not have been possi-

ble without their collaboration.

REFERENCES

Aguiar, J., Portela, F., Santos, M. F., Machado, J., Abelha,

A., lvaro Silva, Rua, F., and Pinto, F. (2013). Perva-

sive information systems to intensive care medicine -

technology acceptance model. In Proceedings of the

15th International Conference on Enterprise Informa-

tion Systems - Volume 1: ICEIS,, pages 177–184.

Boyle-Makowski, R. M. D. (1987). The importance of na-

tive pollinators in cultivated orchads: Their abundance

and activities in relation to weather conditions. In

Proceedings of the Entomological Society of Ontario,

pages 125–141.

Camargo, R. C. R., Pereira, F. M., and Lopes, M. T. (2002).

Sistema de Produc¸ao de Mel. EMBRAPA Meio Norte.

Coelho, J. R. (1991). Heat transfer and body temperature in

honey bee (hymenoptera: Apidae) drones and work-

ers. Environmental Entomology, 20(6):1627–1635.

Dutra, R. and Amorim, C. (2010). Modelo de comunicac¸ao

enderec¸ada por interesses. Technical report, Technical

report, ES 733/PESC-COPPE-UFRJ.

Dutra, R. C. (2012). PhD thesis, Universidade Federal do

Rio de Janeiro.

Dutra, R. C., Granja, R. S., Moraes, H. F., and Amorim,

C. L. (2010). Repi: Rede de comunicac¸ ao enderec¸ada

por interesses. In SBRC’2010, Simp

osio Brasileiro de

Redes de Computadores e Sistemas Distribuidos.

Dutra, R. C., Moraes, H. F., and Amorim, C. L.

(2012). Interest-centric mobile ad hoc networks. In

NCA’2012, Proceedings of the International Sympo-

sium onNetwork Computing and Applications (NCA),

pages 130–138. IEEE.

Esch, H. (1976). Body temperature and ﬂight perfor-

mance of honey bees in a servo-mechanically con-

trolled wind tunnel. Journal of Comparative Physi-

ology, 109(3):265–277.

Giammarini, M., Concettoni, E., Zazzarini, C. C., Orlan-

dini, N., Albanesi, M., and Cristalli, C. (2015). Bee-

hive lab project - sensorized hive for bee colonies life

study. In Proceedings of the International Workshop

on Intelligent Solutions in Embedded Systems, pages

121–126. IEEE.

Goudalo, W. and Kolski, C. (2016). Towards advanced

enterprise information systems engineering - solving

resilience, security and usability issues within the

paradigms of socio-technical systems. In Proceed-

ings of the 18th International Conference on Enter-

prise Information Systems - Volume 2: ICEIS,, pages

400–411.

Heinrich, B. (1975). Thermoregulation in bumblebees.

Journal of Comparative Physiology B: Biochemical,

ICEIS 2017 - 19th International Conference on Enterprise Information Systems

586

Systemic, and Environmental Physiology, 96(2):155–

166.

Hess, W. R. (1926). Die temperaturregulierung im bi-

enenvolk. Zeitschrift f

ur vergleichende Physiologie,

4(4):465–487.

Himmer, A. (1927). Der soziale w

armehaushalt der honig-

biene - ii die w

arme der bienenbrut. Erlanger

Jahrbuch f

ur Bienenkunde, 5:1–32.

ISO (1998). ISO 9241-11: Ergonomic Requirements for Of-

ﬁce Work with Visual Display Terminals (VDTs): Part

11: Guidance on Usability.

Kandjani, H., Andargoli, A. E., Mohtarami, A., and

Shokoohmand, R. (2013). A conceptual framework

to classify strategic information systems planning

methodologies. In Proceedings of the 15th Interna-

tional Conference on Enterprise Information Systems

- Volume 2: ICEIS,, pages 190–196.

Kviesis, A. and Zacepins, A. (2015). System architectures

for real-time bee colony temperature monitoring. Pro-

cedia Computer Science, 43:86–94.

Kviesis, A. and Zacepins, A. (2016). Application of neu-

ral networks for honey bee colony state identiﬁca-

tion. In ICCC’2016, Proceedings of the 17th Inter-

national Carpathian Control Conference, pages 413–

417. IEEE.

Kviesis, A., Zacepins, A., and Riders, G. (2015). Honey bee

colony monitoring with implemented decision support

system. In Proceedings of the 14th International Sci-

entiﬁc Conference on Engineering for Rural Develop-

ment, pages 446–451. Latvia University of Agricul-

ture.

Lent, R. (2008). Improving the location of mobile agents

on self-aware networks. In Proceedings of the Tenth

International Conference on Enterprise Information

Systems - Volume 5: ICEIS,, pages 239–242.

Magno, M., Murphy, F. E., O’Leary, L., Troy, K., Whelan,

P., and Popovici, E. M. (2015). Big brother for bees

(3b); energy neutral platform for remote monitoring

of beehive imagery and sound. In IWASI’15, Proceed-

ings of the 6th International Workshop on Advances in

Sensors and Interfaces, pages 106–111. IEEE.

Murphy, F. E., Popovici, E., Whelan, P., and Magno,

M. (2015). Development of an heterogeneous wire-

less sensor network for instrumentation and analy-

sis of beehives. In I2MTC’2015, Proceedings of

the IEEE International Instrumentation and Measure-

ment Technology Conference, pages 346–351. IEEE.

Noran, O. (2013). Enhancing enterprise information sys-

tems synergy in disaster management. In Proceed-

ings of the 15th International Conference on Enter-

prise Information Systems - Volume 3: ICEIS,, pages

275–283.

Rabaey, M., Vandijck, E., Vandenborre, K., Tromp, H., and

Timmerman, M. (2006). Business process embedded

information systems - for ﬂexibility and adaptability.

In Proceedings of the Eighth International Confer-

ence on Enterprise Information Systems - Volume 1:

ICEIS,, pages 144–150.

Seeley, T. D. (2006). Ecologia da Abelha: Um Estudo de

Adaptac¸

ao na Vida Social. Paix

ao Editores LTDA.

Soares, D. and Amaral, L. (2014). Reﬂections on the con-

cept of interoperability in information systems. In

Proceedings of the 16th International Conference on

Enterprise Information Systems - Volume 1: ICEIS,,

pages 331–339.

Southwick, E. E. (1985). Allometric relations, metabolism

and heart conductance in clusters of honey bees at cool

temperatures. Journal of Comparative Physiology B,

156(1):143–149.

Stabentheiner, A., Pressl, H., Papst, T., Hrassnigg, N., and

Crailsheim, K. (2003). Endothermic heat production

in honeybee winter clusters. Journal of Experimental

Biology, 206(2):353–358.

Winston, M. L. (1991). The Biology of the Honey Bee. Har-

vard University Press.

Zacepins, A., Kviesis, A., Ahrendt, P., Richter, U., Tekin,

S., and Durgun, M. (2016). Beekeeping in the fu-

ture - smart apiary management. In ICCC’2016, Pro-

ceedings of the 17th International Carpathian Control

Conference, pages 808–812. IEEE.

Zacepins, A., Kviesis, A., Durgun, M., and Tekin, S. (2015).

Application of wireless sensor networks in precision

apiculture. In Proceedings of the 14th International

Scientiﬁc Conference on Engineering for Rural De-

velopment, pages 440–445. Latvia University of Agri-

culture.

Zacepins, A. and Meitalovs, J. (2014). Implementation of

multi-node temperature measurement system for bee

colonies online monitoring. In Proceedings of the

2014 15th International Carpathian Control Confer-

ence (ICCC), pages 695–698.

Zacepins, A. and Stalidzans, E. (2012). Architecture of au-

tomatized control system for honey bee indoor win-

tering process monitoring and control. In ICCC’2012,

Proceedings of the 13th International Carpathian

Control Conference, pages 772–775. IEEE.

myBee: An Information System for Precision Beekeeping

587