You are on page 1of 13


Available at


journal homepage:

Management and control applications in
Agriculture domain via a Future Internet Businessto-Business platform
Sokratis Barmpounakis a,*, Alexandros Kaloxylos b, Aggelos Groumas a,
Lampros Katsikas a, Vasileios Sarris a, Konstantina Dimtsa a, Fabiana Fournier d,
Eleni Antoniou c, Nancy Alonistioti a, Sjaak Wolfert e

Department of Informatics and Telecommunications, University of Athens, Greece
Department of Informatics and Telecommunications, University of Peloponnese, Greece
OPEKEPE, Greek Payment Authority of Common Agricultural Policy (C.A.P.) Aid Schemes, Greece
IBM-Research Haifa, Israel
Logistics, Decision and Information Sciences Group, Wageningen University, Netherlands



Article history:

The Agriculture business domain, as a vital part of the overall supply chain, is expected to

Received 9 July 2014

highly evolve in the upcoming years via the developments, which are taking place on the

Received in revised form

side of the Future Internet. This paper presents a novel Business-to-Business collaboration

28 April 2015

platform from the agri-food sector perspective, which aims to facilitate the collaboration of

Accepted 30 April 2015

numerous stakeholders belonging to associated business domains, in an effective and flex-

Available online 14 May 2015

ible manner. The contemporary B2B collaboration schemes already place the requirements
for swift deployment of cloud applications, capable of both integrating diverse legacy sys-


tems, as well as developing in a rapid way new services and systems, which will be able to

Future Internet

instantly communicate and provide complete, ‘‘farm-to-fork’’ solutions for farmers, agri-


food and logistics service providers, ICT companies, end-product producers, etc. To this

Generic Enablers

end, this conceptual paper describes how these requirements are addressed via the FIspace

B2B platform

B2B platform, focusing on the Greenhouse Management & Control scenarios.


2015 China Agricultural University. Production and hosting by Elsevier B.V. All rights

Internet of Things



ICT and agriculture originate from disparate human needs,
nevertheless, the first domain proves to be of upmost

significance to the second in order to facilitate modern complex business processes related to agriculture. ICT application
in agriculture has been an emerging field for some years,
attempting to enhance the agricultural processes through
sophisticated information and communication developments. Aspects of the agriculture industry such as crop cultivation management and control, quality management,
transport of food products and food preservation may all be
enhanced by taking into account their domain-specific
requirements and translating them into the respective functional design, development and applications by ICT experts.

* Corresponding author at: Department of Informatics and
Telecommunications, National & Kapodistrian University of
Athens, Panepistiomopolis, Ilissia, Athens 15784, Greece. Tel.:
+30 2107275176.
E-mail address: (S. Barmpounakis).
Peer review under the responsibility of China Agricultural
2214-3173  2015 China Agricultural University. Production and hosting by Elsevier B.V. All rights reserved.

a considerable amount of efforts is focusing on wireless sensor networks solutions for monitoring the condition of the crops during production. such as the aforementioned ones. In [11]. In the afore-mentioned work. along with a framework that allows the interconnection among services developed by different service providers. which is capable of bridging the diverse gaps between different – but directly associated – domains. involving more and more business actors and stakeholders. Future systems cannot rely on existing infrastructure and require more sophisticated and smart frameworks in order to improve the effectiveness of business processes.e. Current trend is gradually heading toward the Internet capabilities in order to address several of the existing limitations.. Several sophisticated Farm Management Information Systems’ solutions and architectures have been already described in the literature ([6–8]). enterprises. Farm Management Information Systems (FMIS). although there is a single platform to support the several apps (Android or iOS respectively). knowledge inference and finally knowledge transfer back to the farm or greenhouse and execution of the generated actions. Another FI-PPP Phase 1 project related with the Smart Agri-Food project presented earlier. actual execution. all are terms. The increasing complexity through the agri-food supply chain processes. precision agriculture. is ultimately being able to offer to the end-users (i. what is actually missing is a holistic. in a similar manner the famous Apple Store ([16]) or Google Play ([17]) offer. Smart farming. which aims to .52 Information Processing in Agriculture Several efforts over the last years have attempted to integrate end-to-end ICT solutions into the agriculture business processes. In [13] the implementation of an innovative. however. previously unexploited. namely an ‘‘app store’’. farmers. which introduce the notion of sophisticated processes using advanced information and communication tools and systems into agriculture processes. In [9] the authors describe a novel architecture. weather. farm business apps. With regard to existing solutions. An app store usually offers a variety of choices among apps. as well as external back-end systems. resource coordination. an advanced architecture based on the Future Internet developments is presented. The main operations. which is processing the available data and generates actions – often in an automated manner – for the farmer. using dynamic. Even in the case of Google Play or Apple Store. the apps are developed in a completely independent and separated way from one another. Analyses of potential developments in the precision agriculture domain have taken place taking into account spatial and temporal variations. usability and user friendliness. AgWeb app store ([18]) offers a considerable number of agrifood chain-related apps specializing in different fields such as agronomy apps. services) to accomplish it. greenhouse managers.e. data processing. the proposed schemes attempt to partially enhance a narrow aspect of the overall supply chain picture. The aim of FInest was to define realistic business scenarios illustrating how transport business operations could be conducted and facilitated through the help of a FI-based collaboration platform. end-to-end solution. machinery. as well as select and deploy the desired means (apps. Examples of existing site-specific precise farming approaches.) and/or domain business stakeholders the capability to search for and use specific services and applications according to own needs and requirements. An operational example is used to demonstrate the interworking of the functional modules of the architecture. real time. However. In the agriculture domain. adjustable approaches as well as future directions for such developments have been described ([1. not being able to maximize the potential outcomes. as well as a processing component. which takes advantage of the Future Internet Public–Private Partnership (FI-PPP) ([10]) capabilities in order to facilitate the interoperability among services and stakeholders. Different use case scenarios were used to demonstrate the capabilities of the FInest architecture in real-life: handling late booking cancellations. In this paper. most of the existing schemes limit their functionality to a very narrow aspect of the overall business process. Internet of Things domain). However.. Another significant perspective. which are in general dealt within the context of Precision Agriculture and Greenhouse Management are data monitoring. while at the other end there is a database. cloud-based Farm Management System is provided. lacking any business collaboration model behind them to link their requirements. such as an expert system. etc. real-time event handling. open and flexible. from a centralized repository of services/apps. a detailed analysis of the FI-PPP software modules exploitation is provided according to the deployments in the context of the Smart Agri-Food (SAF) FI-PPP Phase 1 project ([12]). handling vast number of devices (i. lacking this way in homogeneity. as well as realizing an efficient communication scheme between the involved stakeholders of the business chain still remain some of the most significant obstacles. In principle. as well as corresponding business actors and events. which supports the execution of the offered apps. The complexity of those business processes is nowadays the most crucial challenge that needs to be coped with. failing to provide a holistic solution. for product autoidentification. eplanning. Modern business processes related to agriculture and greenhouse management and control involve multiple stakeholders. etc. based on the FI 2 ( 2 0 1 5 ) 5 1 –6 3 technologies as well but focusing on the domains of transport and logistics is FInest ([15]). is a unified underlying infrastructure. What is actually missing. as well as for the transport of products condition monitoring ([3–5]). in parallel to the actual developments and technology choices presented earlier. it is aimed to demonstrate how the adoption of these general-purpose software modules provided by FI-PPP and their extension into farming specific ones may provide a cloud operating system that can integrate different services and applications. In all cases presented earlier. Although prior efforts do attempt to address some of the high challenges of the business processes either in the domain of the agriculture or the logistics exclusively. a typical solution comprises a wireless sensor network residing inside the farm or greenhouse. in which any stakeholder is able to take advantage of a specific business collaboration model. as well as automated shipment tracking. as well as multiple and often complex interactions between them establish completely new needs – previously unseen. or results and offer a complete end-to-end solution.2]). isolating the business actors and thus.

etc. 3. Repository GE etc.1. in order to facilitate 2 ( 2 0 1 5 ) 5 1 –6 3 53 the creation of novel applications built on top of the existing infrastructure features. etc. searchable. start and accomplish collaborations.. which engages both developers and entrepreneurs and provides the required tools to support it..g. The Future Internet Public–Private Partnership (FI-PPP) and FI-WARE Generic Enablers In the attempts discussed earlier. 1.): Enables the implementation of modern FI applications. 2. Internet of Things GEs (e. Business Calculator. comparable to Facebook or LinkedIn. accessible. energy etc. apps’ repositories. Cloud Hosting GEs (e. Data/Context GEs (e.): Enables the exchange and publication of massive data in a fast. tourism. Through cloud solutions. DB Anonymizer. flowers and plants supply chain monitoring and meat information provenance. demonstrating how the FIspace platform is exploited in order to facilitate B2B collaboration scenarios related to the agricultural.. manufacturing. The FIspace platform and the FIspace operational model 3. and allowing seamless collaboration between multiple business actors. Identity Management GE. fish distribution and (re-)planning. fruit and vegetables quality assurance. while diverse domain-specific apps and services are deployed and configured in accordance with these business entities. we summarize the conclusions. integrated and extensible ‘‘Platform as a Service’’ (PaaS) for stakeholders and business players belonging to diverse business domains. General-purpose software modules – known as ‘‘Generic Enablers’’ (GEs) – support the infrastructure of FIWARE. Device/Network Interfaces GEs (e.g. The main idea of FIspace is that any business collaboration is ‘‘built’’ around ‘‘business entities (BEs)’’. logistics. In Section 4. Often. the FI-PPP plays an essential role toward creating the required communication and technological infrastructure for supporting a complex and holistic stakeholder interaction in the context of business collaboration processes. The rest of the paper is organized as follows. etc. Gateway Data Handling. – an ecosystem.): Helps the ‘‘things’’ connection to the real world by becoming available. which collaborate in order to fulfill a complete business case: FIspace ([19]).g. Furthermore. These GEs are the building blocks of the core platform and are categorized according to their principal scope to several categories such as: • • • • • • Applications/Services Ecosystems GEs (e. and usable for FI uses and services. Section 3 provides an overview of the FIspace platform. FIspace Platform high-level architecture FIspace is a Business Collaboration Network in agri-food.g. . safety.g. Future Internet refers mainly to an initiative. which will be described in the following sections. eHealth.. Object Storage GE. smart applications deployed inside FIspace.g. providing cloud-hosting capabilities for third parties in order to be able to experiment with Future Internet applications. the FIspace operational model) that can be seen as a kind of handbook of how this reference architecture could be implemented. Business can interact with each other. The vision of FI-PPP is based on FIWARE and FI-LAB ([14]). and treating any kind of information as objects. it aims to create a core platform consisting of various ‘‘sub-platforms’’ and sub-modules. Application Mash-up. a reference architecture for Business-to-Business (B2B) collaboration in supply chain networks is described including a methodology (i. Cloud Edge. greenhouse management and control. Complex Event Processing. we focus on the greenhouse management and control trial. which performs research activities to create novel architectures for the Internet. Although. In Section 2. FI-PPP mainly targets toward the digitization of the European economy sectors including agriculture. The high-level architecture of FIspace platform is demonstrated in Fig. based on the Future Internet technologies. focusing on multiple business domains: crop protection information sharing.. compliant with cloud hosting infrastructure.. smart cities. Big Data Analysis. semantic tools or big data handling.): Enables the implementation of the required security mechanisms for each app to be developed. the user can select among more than one implementation of a Generic Enabler is available from different providers. in order to evaluate specific key performance indicators. independent of their storage or location. Security GEs (e. Network Information and Control GE. Various use case scenarios are applied in the context of FIspace. In the final section. It may be described as a business-to-business software tool. via the FI-LAB environment Generic Enablers implementations are available and globally accessible as services. the current work emphasizes on the greenhouse management and control use case trial of the FIspace project. IoT Broker GE.): Establishes an open and standardized network infrastructure. ranging from small. transport and logistics aiming to provide an innovative business space.Information Processing in Agriculture provide an end-to-end. FI-LAB is a use case example of FI-WARE instance. which are derived and insights on the next steps are provided.): Supports the creation of an innovative ecosystem of applications and services. exchange information. the platform supports multiple domains in order to support the complete supply chain ‘‘from farm to fork’’. Research activities that could be seen as components of a future Internet include network management. while diverse terms and conditions are applied among the various GEs for external use. network virtualization. and how all the involved agri-domain business actors may benefit from such a B2B platform via novel. Each one of the GEs is licensed with no costs within the FI-PPP program. etc. etc. yet efficient way. Access Control.e. Semantic Annotation. a description of the Future Internet PPP developments is presented. communicate and coordinate activities. incremental steps to complete architecture and principles’ redesigns. Job Scheduler. transport.

Internet of Things) in order to communicate with the FIspace platform and participate in a business collaboration. and monitor collaborative processes in the FIspace platform. Fig. access to FIspace App Store. The FIspace Store: the access point from which the FIspace apps are available for provisioning. notifications mechanism. To this end. on which the business collaborations rely on. Identity Management and Revenue Sharing GEs ([22–25]). handle and manage the FIspace collaboration processes [28]. The BCM component is responsible to orchestrate the different processes from different stakeholders and assure the According to the previous section. The FIspace Cloud Service Bus (CSB): another essential core module. several GEs are already used. Privacy and Trust (SPT) module: it is concerned with all the essential features in order to provide all the required security mechanisms in order to ensure a reliable exchange of business information and transactions. The Security. 2 – FI-WARE FI-LAB’ GEs and FIspace Specific Enabler. The System and Data Integration Module is using the Mediator GE ([27]). integrated user authentication mechanism etc. consumption and purchase both for consumers as well as the FIspace app developers. It is comprised of o the Business Collaboration Module (BCM): this module serves the principal real-time B2B collaboration concept. o the Event Processing Module (EPM): A complex event processing engine that monitors and tracks after events of interest related to the business collaboration.2. The B2B Collaboration Core module: it is one of the core components of the platform. As it is illustrated in the architecture above. reputation and recommendation mechanisms. 3. which provides all the essential features for the business collaboration. 1 – FIspace high-level platform architecture. • • The FIspace Store is built upon the WStore GE ([21]). which facilitates the interoperability among different communication protocols and data models. and this number may also increase as the developments progresses: • Part of the FIspace Front-End is based on the Wirecloud GE ([20]). taking care of the secure access to the FIspace networks. execute. Privacy & Trust Module is based on the Identity Management GE mentioned earlier. The B2B collaboration core The aim of this module is to create. being one of the Specific Enablers example. etc. an ‘‘application mash-up’’ platform. . 2) illustrates the relationship between FI-WARE developments. The Event Processing Module (EPM) core module implements the Complex Event Processing (CEP) GE ([26]). Registry. two complementary components are implemented: The Business Collaboration Module (BCM) and the Event Processing Module (EPM).g. manage. which facilitates the connecting between all the main components of the platform. Marketplace. as several of the core components of the platform is relying up to an extent to a generic enabler. the platform is comprised of six fundamental modules: • The Graphical User Interface (Front-End): the main access point for the end-users. its purpose is to create. FI-LAB.54 Information Processing in Agriculture 2 ( 2 0 1 5 ) 5 1 –6 3 Fig. which respectively connects to Repository. the FIspace platform is acting as a ‘‘specific enabler’’ of FI-WARE. • • • • • • • The System and Data Integration module: the ‘‘gateway’’ of any external or legacy systems (e. Generic Enablers and FIspace platform. The figure that follows (Fig. Although the platform development is still ongoing. The Security.

i. Guards are defined as conditions of the form ‘‘on <event> if <condition>’’ (visually represented as diamonds). that is. An entity type includes both a data schema and a lifecycle schema.e. the analysis of events coming from many sources. we can distinct between situations that result from the actual execution of the process or collaboration and situations that result from external events (i. Each Stage can own one or more Guards. In FIspace we apply the GSM – Guard. FIspace operational model The business collaborations within FIspace platform heavily rely on the respective business entities. Stage. a delay in a delivery etc. a sensor reading outside a permitted range. Reactive rules analyze past events and derive situations by applying pattern matching over a single or a set of events over time. 3. 3. This is accomplished through the online forecasting of future events. on the other hand. Examples of situations of interest can be: missing documentation at a certain point in time. In order to realize the collaboration among business actors and take advantage of the business entities.e. The BCM is based on the entity-centric approach [28].3. They represent the achievement of distinct business objectives and are visually represented as circles.Information Processing in Agriculture 2 ( 2 0 1 5 ) 5 1 –6 3 55 Fig. Proactive rules. The EPM processes these events and by applying pattern matching derives situations of interest ([29]). In general. as business entities. which are tightly linked. external systems. The events sources (aka event producers) can be the actual execution of the collaboration (i. In the Greenhouse scenario in FIspace we defined Advice as our key business entity related to the Greenhouse Management & Control scenarios to demonstrate in a more specific way how a GSM model may define the greenhouse management workflows processed by the FIspace platform. situations that require appropriate actions. The lifecycle schema of an entity type specifies the different ways that an entity instance might evolve as it moves through the overall process. apps are developed and available to the FIspace users via the FIspace Store. Stages represent clusters of activity and are visualized as rectangles with rounded corners. relate to situations that are likely to happen in the (near) future. events coming from external systems or sensors). The apps are built within a generic context and are not supposed to be linked to specific external systems. acting internally. This approach relies on the notion of entities (aka. Milestone-model ([28]). 3 – GSM model for describing the lifecycle of a business entity. or capitalize on predicted opportunities – in advance. Stage can have one or more Milestones. Like Guards. or sensors. artifacts. the BCM). these business entities reside within the platform. The data schema provides an end-to-end conceptual view of the key data for this entity type. and the application of online decision-making processes ([30–33]). correct sequence of the tasks execution. However. . The EPM can support two types of situation detection capabilities: reactive and proactive. which were described in the previous sub-section. GSM is a declarative model that enables hierarchy and parallelism of the tasks and reactive in nature. In order for them to be usable. as a result the end-users are not supposed to interact with them directly. which control its activation. In general. A business entity is a key concept that evolves as it moves through a business process. the tasks flow as a result of triggering events to specific conditions. we refer to proactive event-driven computing as the ability to mitigate or eliminate undesired states. or dynamic artifacts). The main constructs of the GSM model are illustrated in Fig.e. they are noted in the form ‘‘on <event> if <condition>’’. The Event Processing Module (EPM) component monitors events and detects situations of interest. the business user needs to instantiate (configure) them according to own needs.

. agronomists. assume there is a business collaboration between a greenhouse owner (or farmer). as well as an expert (advisory) system. advisory services enterprises. using probably notifications mechanism. different scenarios take place. 4 below illustrates the connection between the apps and the FIspace business entities. several generic as well as domain-specific applications are developed according to the platform’s operational model. performance. Fig. which will consume events produced within this business collaboration. a third application could be deployed in order to receive from the expert system the advised actions and provide them in a proper way. in the context of which. Farm Management Information . The scope of the experiment is the management and control enhancement of greenhouse operations. mapping specific external systems for a respective specific instance of the business entity.: farmers. greenhouse managers.e. In the context of the Greenhouse Trial. similar to this one. a second app could illustrate the weather information provided by the weather service. These scenarios are summarized as follows: • • • • Requesting for greenhouse advice from an expert system Managing consumer complaints Facilitating new collaborations between stakeholders Recalling a product using traceability mechanisms For each one of the above scenarios. end-product producers.e. increases the agricultural productivity and revenues as well overall facilitates the several processes. The Greenhouse Management & Control scenarios Real-world business scenarios are established in order to evaluate the FIspace platform. The above process describes the fundamental operational model of FIspace. to the farmer. 4 – The FIspace Operation Model between the business entities and the FIspace apps. an app. in order to demonstrate numerous aspects of the domain that may be enhanced using the FIspace infrastructure. A business entity may be linked to multiple FIspace apps. will be presented in detail in the next section. For instance. which provides actions to be taken inside the greenhouse to the farmer according to the environmental conditions (provided by sensors inside the greenhouse). there is the possibility for one FIspace app to be linked and consume events from more than one business entity. Finally. also focusing on the transport and logistics domain This paper focuses on the ‘‘Farming in the cloud’’ theme. a number of diverse business stakeholders are participating. The purpose of these implementations is to test whether the underlying infrastructure is able to deliver the required functionality. i. privacy and reliability for the business stakeholders of the agricultural domain (in the case of the Greenhouse trial scenarios) to rely on and expand their activities using the FIspace technologies. aiming at demonstrating how the Business-toBusiness collaboration among the domain stakeholders via the FIspace features. focusing on the transport and logistics domain ‘‘Smart distribution and consumption’’. The use case trials primarily originate from two business domains. In addition. Respectively. security. they are categorized along three ‘‘themes’’: • • • ‘‘Farming in the cloud’’. and in particular on the greenhouse management and control use case trial. agricultural domain and transport and logistics. A scenario. i. a weather service provider. FIspace could support multiple apps linked to this particular collaboration: Firstly. focusing on the agricultural domain ‘‘Intelligent perishable goods logistics’’. Notably. which shows the conditions inside the greenhouse could be developed.56 Information Processing in Agriculture 2 ( 2 0 1 5 ) 5 1 –6 3 Fig. 4.

The end-user receives the actions from the expert system via the respective FIspace app. Such a business process. sensor values detected out of predefined boundaries) a request for actions is sent to the advisory system. and the Advisory Service. 5): the farmer/greenhouse manager and an advisory/expert system enterprise. as well as a product traceability platform.) and create new Service Level Agreements in a much more efficient as well as sophisticated way. will also gain profit from participating in such collaborative chains • 2 ( 2 0 1 5 ) 5 1 –6 3 57 Appropriate information that is related to farmers’ profiles could be exchanged automatically with states’ and information policies enabling to simplify daily routines. What is going to be tested. as well as the quality and the Fig. Overall. The message sequence chart (MSC) that follows. which corresponds to respective business collaboration and provides the information via the UI to the farmer. health hazards must be identified and relevant stakeholders must be notified in a timely manner. and afterward they are forwarded to FIspace. The main blocks in the technology layer are FIspace. all of which will collaborate in order to produce the best possible feedback to the stakeholder. it is expected that: • • • • • Farm managers/Farmers will be able to manage their greenhouses in a more efficient way. the interconnected advisory services must inform via the FIspace Domain Specific Apps the farm managers of forthcoming alerts and actions to be taken. which is illustrated later in the paper. targets to illustrate a radical enhancement. . a weather service. weather services. For example. the business layer of the collaboration is illustrated (yellow color). State Agencies etc. an agricultural state agency. maximizing their products’ usage and thus. This specific business collaboration is handled by the advice business entity. revenues Developers will be able to upload via the FIspace Store their apps. where they are contextualized. which will be presented below. which will be used by the various business collaborations Various business actors who will be involved in the different scenarios like Consulting Firms. the Greenhouse FMIS. Inside the FIspace platform an ‘‘Advice Request’’ app resides. In the archimate model above. is whether and to which extent. retrieve updated information about products and be notified for any emergency situations concerning their products Legacy/Back-end systems’ owners will deploy their systems in multiple collaboration chains. In case of events inside the greenhouse (i. via the use cases. The idea is that the sensors’ values of the Greenhouse are forwarded to the Greenhouse Farm Management System (FMIS). Requesting for greenhouse advice from an expert system Two main business actors are involved in the first scenario (Fig. end-product producers etc. as well as the technology/infrastructure layers (gray layer at the bottom). in principle with regard to the time required accomplishing it. handle their tasks more efficiently retrieving information from multiple back-end systems. Greenhouse domain stakeholders must be able to discover new potentials for collaborations (with farmers. advisory systems. 6). which provides advisory services to the greenhouse based on the conditions inside the greenhouse.Information Processing in Agriculture System (FMIS) owners. which will potentially enable him to maximize his revenues End-Product producers will be able to discover potential partners that interest them in a much more efficient way.e. a direct link between a farmer and the state agency would eliminate bureaucracy. 5 – Collaboration (business and technology layers) for Advice Request from expert system for actions to be taken inside the greenhouse.1. certain Business KPIs are improved. although being a simple business case by involving only two stakeholders. describes the particular process (Fig. The farmers must be able to organize the task planning of their greenhouses in an automated and more sophisticated way. 4.

a back-end system provider as well as a Consulting Firm. 7). This will result. Notably. on the one hand the farmer. The Message Sequence Chart. when trying to trace food quality issues created during the supply chain. Similarly. accuracy of the actual advice actions. Instead of having to send separately the complaint report to each one of the aforementioned business actors. 8). 7 – Collaboration for Complaint Management from an end-product producer. 4. Managing consumer complaints In the Managing Complaints scenario (Fig. further elaborates on the particular scenario and the interaction of the diverse systems and the FIspace platform: Fig. . consequently. The Managing Complaints collaboration is managed inside the platform by the complaint business entity. Indicatively. to higher profits for all involved stakeholders. a weather service could be linked to the platform – using the respective adapter – and provide information to the advisory service. and – based on the respective business entity configured inside the platform – forwards the appropriate requests accordingly to the required systems. who based on the quality of the services will have the possibility to increase the collaborations with interested parties. the specialized FIspace app receives the complaint. the concept of the platform architecture and operational model provides the opportunity for further enhancing the outcome of the particular business collaboration by deploying more business players and external systems to participate in the same business entity. which follows (Fig.58 Information Processing in Agriculture 2 ( 2 0 1 5 ) 5 1 –6 3 Fig. 6 – MSC for the Greenhouse Advice Request scenario.2. but on the other. the advisory system owner as well. so that ultimately the information forwarded back to the farmer is enhanced. a logistics service could connect to the existing business entity in order to get updates with regard to the conditions inside the greenhouse. an end-product producer receives a complaint from a consumer. The end-product producer uses the Managing Complaints FIspace app to submit the complaint and receive a respective analysis on it from a traceability service (TS) provider.

3. In the message sequence chart that is presented below (Fig. Similarly with the previous scenarios presented above. availability.g. in order to provide back information. while the processed product’s part to the traceability platform. 8 – MSC for the Complaint Management scenario. which need to be notified for the threat. which the developed FIspace app will use to extract the list of all the involved parties. which are used for storing the service information in the RDF semantic web standard [34]. The State Agency is using the FIspace Product Recall App and sends a request after submitting the relevant information to discover the users of the hazardous pesticide. In addition to the above. multiple parameters need to be taken into account (location.. an automated match-making mechanism between available offers and demands is available. the MSC of the specific use case is demonstrated in Fig. 10 below: An example of an initial creation of a service offer by user A and a service demand by user B sometime later is illustrated. as well as forwards the query to external systems. the location. as well as FMIS from several greenhouses are also connected to the business process. any business stakeholder is able to select among the domain of preference (e. the farmers’ database of a State Agency of Agriculture is used. Facilitating new collaborations between stakeholders Discovering the ideal business players to collaborate with for accomplishing diverse business processes is not a simple task. Many e-marketplaces and/or databases can ultimately integrate with the Marketplace Operations service of the platform. which may possibly satisfy their requirements.Information Processing in Agriculture 2 ( 2 0 1 5 ) 5 1 –6 3 59 Fig. 9).e. 11) it is assumed that an emergency event breaks out and needs to be dealt with in a timely manner. the platform’s ‘‘gateway’’) to the respective external systems: the information regarding the raw product part of the request will be forwarded to the Farm Management Information System of the greenhouse. The traceability service (TS).e. by the time an end-product producer enterprise begins to search for new farmers-suppliers for example until an SLA is established) may take very long time. The match-making mechanism within FIspace provides an instant notification to both users for informing them so that they possibly proceed to the creation of a new collaboration. Often. The existing offer created by user A satisfies the demand. Recalling a product using traceability mechanisms In this scenario (Fig. the period of the collaboration in which one is interested and directly make a request to FIspace. which has been linked to the business entity. The Business Collaboration core module receives the request from the Complaint Management app’s back-end and dispatches the request via the System and Data Integration (SDI) module (i. The platform – via the Marketplace Operations – is querying the already created service offers or demands. in the particular scenario (Fig.4. specialized services etc. which is submitted by user B. 4. which via the app’s respective front-end provides the required info to the user. a health threat is discovered by a State Agency for agricultural policies due to a hazardous pesticide.) and the whole process (i. via the SDI module the information is aggregated inside the Business Collaboration module and transferred back to the back-end of the Complaint Management app. in order to notify the interested parties for existing services. which have been integrated and connected to FIspace. agriculture – greenhouse products). Using the Marketplace Operations service within the FIspace platform. 12). providing the highest-quality possible services to the interested business partner. The implementation of the Marketplace Operations back-end is heavily relying on two GEs from FI-WARE: the Repository GE and the Marketplace GE. Similarly. 4. the process is initiated by a State Agency request . More specifically.

60 Information Processing in Agriculture 2 ( 2 0 1 5 ) 5 1 –6 3 Fig.5. 9 – Facilitating the search for new collaborations using the FIspace platform. The following messages involve the information acquisition from the external parties (Traceability platform and Greenhouse FMIS) and the final feedback that is provided back to the State Agency: As it is illustrated in the above figure. . 10 – MSC for the New Collaborations scenario. toward the FIspace platform. the GSM model of the BEs for two of Fig. Event Processing and Business Collaboration modules is in charge afterward of the overall coordination of the flow of the messages and the final notification to the user via FIspace’s front-end. the request is initially submitted via the SDI module into the FIspace platform.e.. Greenhouse trial business process modeling via the notion of business entities In order to provide further insights to the operational model of the FIspace platform regarding the business entities (BEs) of the greenhouse trial. i. A tight collaboration between two core modules of the platform. 4.

We highlight in Table 1 below the main steps of one of the two GSM models (Advice BE – Fig. These BEs are linked to the apps residing inside the platform. Respectively. The first figure (Fig. 14) shows the BE lifecycle (GSM model) of the Managing Complaints use case. and generated by the expert system. 13 and 14). the second one (Fig. Fig. 11 – Collaboration for recalling a product from Agricultural State Agency. which were presented in the technology layer of the respective scenario. 12 – MSC for recalling a Product scenario. 13). the afore-presented scenarios (namely ‘‘Requesting for Greenhouse Advice from an Expert System’’ and ‘‘Managing Consumer Complaints’’) is illustrated below (Figs. requested by the FIspace platform. 13) illustrates the advice business entity lifecycle (GSM model) for the greenhouse. The BEs are: Advice BE (represented by the Advice GSM) and ComplaintAnalysis BE (represented by the Complaint Analysis GSM) respectively. Fig. in order to demonstrate the role of the business entity in the lifecycle of the collaboration process.Information Processing in Agriculture 2 ( 2 0 1 5 ) 5 1 –6 3 Fig. 61 . 13 – Advice GSM.

as the complexity of future collaborations will require diverse parties participating. a critical perspective that needs to be highlighted is the facilitation of integrating existing legacy systems (e. This is realized by minimizing the required effort to harmonize the communication schemes between the platform and these systems. Scenario steps • • • • Greenhouse sensor values are being constantly monitored by the EPM. etc.onAchieve() guard is passed and NotifyFarmer is activated FarmerNotified is reached – AdviceHandled is reached – HandlingAdvice stage ends Furthermore. FMISs. A B2B collaboration core module is used for the orchestration of such business processes. Discussion and conclusions In this conceptual paper we presented how FIspace. an innovative B2B collaboration platform.62 Information Processing in Agriculture 2 ( 2 0 1 5 ) 5 1 –6 3 Fig. From the agri-food domain’s perspective in particular. 14 – Complaint Analysis GSM. via the greenhouse management and control trial’s developments that were provided. are detected and as a result the EPM emits the OutOfBoundariesNotification event to the BCM Expert system responds back with required advices/ actions after receiving the values The expert system notifies the FIspace platform that the advice is given to the greenhouse FMIS A notification is created inside FIspace platform for the farmer. making FIspace one of the cornerstones of the forthcoming Business-to-Business technological solutions in the near future. Advisory Systems. and more specifically the RequestingAdvice sub-stage SendRequestForGreenhouseAdvice action is activated AdviceGiven milestone is reached upon successful receiving the requested Advice The AdviceGiven. At some point in time one or more sensor values that exceed pre-defined thresholds. This harmonization is achieved via another crucial module of the B2B scheme.g. visible upon entering FIspace and opening the Advice app Of course. more complex collaborations can be accomplished by combining BEs along with their respective GSM models. set the foundations to establish Future Internet-enabled software tools. the System and Data Integration module of the platform. . as well as huge number of events that need to be handled appropriately. as well as the operational model of the platform that were also presented in detail. multiple data types and standards being used for the accomplishment of a single collaboration. Table 1 – Request for Advice from Expert System scenario.). onEvent() is received and the process enters the HandlingAdvice stage. The architecture. Respective progress of the Advice BE GSM model • • • • • An OutOfBoundariesNotification. will reshape modern business collaborations with the help of the afore-presented notion of the business entities. 5. a completely novel way of linking diverse business stakeholders and external systems – previously completely disconnected – is presented.

fispace. [9] Kaloxylos A. 2014. In: Proc. Brewster C. A basic model for proactive event-driven computing. San Jose. [16] Official Apple Store. Corfu identity-management-one-idm. Groumas A. DEBS ‘11 Proceedings of the 5th ACM conference on Distributed Event Based systems. Wolfert S. [17] Google [11] Kaloxylos A. 2007. IEEE Pervasive Comput. DEBS ‘12 Proceedings of the 6th ACM conference on Distributed Event Based Systems. Electron. Etzion O. Available from: app_store. SRII ‘12 Proceedings of the 2012 Global Conference. Sarris V. Ancev T. 2013. Engel Y. 2014. Fournier F. Available from: http://www. [21] WStore GE. Deljoo Fountas S. Event processing in action. Saraiva M. et al. Comput.smartagrifood. USA. Springer. Greece. CA. . vol. In: Proc. [26] Complex Event Processing (CEP) – IBM Proactive Technology. Comput.3(1):38–45. Noor A. p. [28] Hull R. HAICTA ‘13 Proceedings of the 6th International Conference on Information and Communication Technologies in Agriculture. [15] FInest – Future Logistics. 2014. Singh R. et al. [18] AgWeb App Store. Damaggio Robbemond R.2(8):3955–63. Available from: http://www. 313– 22. 2014. 2014. [8] Sørensen CG. Metzger A. USA. 2012. Precision Farming: Challenges and Future Directions. Fournier F. FI-WARE Catalogue. Int. Software architecture for farm management information systems in precision agriculture. Comput. 2011. Verwaart T. Shrank C.agweb. org/enablers/complex-event-processing-cep-ibm-proactivetechnology-online. [29] Etzion O. Available from: http://www. Introducing the Guard-Stage-Milestone Approach for specifying Business Entity Lifecycles. Franklin Terol CM. Available from: http://store. [24] Registry SAP RI.72(1):37–47.finest-ppp. 2010. 2014. The use of Future Internet technologies in the agriculture and food sectors: integrating the supply chain. Available from: http://www. Electron. Agric. ICSC ‘04 Proceedings of the 2004 (4th) International Crop Science Congress. FI-WARE Catalogue. Arlington. 2014. Heath III F. [5] Baggio A. 2014. [25] Identity Management One – IDM.w3. 2012. 2010. 51– has received funding from the European Commission’s Seventh Framework program FP7-ICT-2012 under grant agreement N 604123 also referred to as FIspace (Future Internet Business Collaboration Networks in Agri-Food. Springer Precision Agric. Teye us. Bouma application-mashup-wirecloud. Comput. Antoniou E. p. Cook SE. Cugnasca E. In: Proc. 2014. Agric. 2014.6(1):7– 6551(1). Agric. Hirakawa R. [12] Smart Agri-Food FI-PPP Phase 1 project. Available from: http://catalogue. Wireless sensor networks in agriculture: for potato farming. p. An infrastructure for the development of distributed service-oriented information systems for precision agriculture. USA. J. 2014. [7] Murakami E. Bochtis Available from: https://www. Ribeiro M. Predictive monitoring of heterogeneous service-oriented business networks: the transport and logistics case. 2004. leading to these results. FI-PPP Use Case Project. Available from: http://catalogue. 2014. p. Whelan B. An efficient wireless sensor network for precision agriculture. 2012. R E F E R E N C E S [1] Dobermann A. [10] Future Internet Public–Private Partnership. Hobson S. NY. Eigenmann R. Nash E.89(1):130–44. Seilonen I. Wolfert J. Magdalinos P.70(2):328–36. [19] Future Internet Business to Business Collaboration Networks in Agri-Food. FI-WARE Catalogue. Electron. et al. p. DEBS ‘13 Proceedings of the 2013 Conference on Distributed Event Based Systems. Eng. Pesonen L. Technol. Adamchuk VI. Texas. 19. Niblet P. [33] Feldman Z.100(1):168–79. 2014. 2014. Feldman Z. Blackmore S. Available from: 2005. Lecture Notes in Computer Science. Available from: http://catalogue. 125–36. et al. Politopoulou Z. Farm management systems and the Future Internet era. 2011. 2010. A cloud-based Farm Management System: architecture and implementation. et al. p. In: Proc. Transport and Logistics). Conceptual model of a future farm management information FI-WARE. 2 ( 2 0 1 5 ) 5 1 –6 3 63 [14] FI-WARE API. Available from: http://catalogue. Food and Environment. Available from: http://catalogue. [22] Repository SAP RI. 97–106. Transport and [31] Engel Y. [34] Resource Description Framework (RDF). Proactive event processing in action: a case study on the proactive management of transport processes. Towards proactive event-driven computing. Australia. Etzion Correa [30] Engel Y. In: Germany. Koskinen K. Electron. Wireless sensor networks in precision agriculture. Brisbane.Information Processing in Agriculture Acknowledgment The research. FI-WARE FI-WARE Catalogue. Future directions of precision agriculture. FI-WARE Catalogue. In: Proc. Manning Publications. Katsikas Gupta M. Comput. 2014. 2014. Pedersen SM. [3] Keshtgary [2] McBratney A. Available from: http:// www. Can. [32] Metzger A. 1–24. Sci. [20] Application Mashup – Wirecloud. Available from: http://catalogue. Agric. [4] Shinghal K.3(1):1–5. Neth. Available from: http://www. [13] Kaloxylos A. FI-WARE Catalogue. 135–46. Multimedia Wireless Netw. p. Franklin Available from: http://catalogue. Srivastava N. J. 2014. [6] Nikkila¨ R. [27] Mediator GE. [23] Marketplace SAP RI.–48. Electron.