欢迎访问一起赢论文辅导网
SCI期刊论文
当前位置:首页 > SCI期刊论文
Big Data in Smart Farming – A review
来源:一起赢论文网     日期:2019-03-03     浏览数:184     【 字体:

 Big Data in Smart Farming A reviewSjaak Wolferta,b,,LanGea, C or V e rd ou wa,b, Marc-Jeroen BogaardtaaWageningen University and Research, The NetherlandsbInformation Technology Group, Wageningen University, The Netherlandsabstract article i nfoArticle history:Re ce iv ed 2 A ug us t 2 016Received in revised form 31 January 2017Accepted 31 January 2017Available online 7 February 2017Smart Farming is a development that emphasizes the use of information and communication technology in thecyber-physical farm management cycle. New technologies such as the Internet of Things and Cloud Computingar e e xp e cte d to le ver a ge th is de vel opm ent and i ntr od uce m or e ro bot s and ar tifi cia l i nt el li gen ce i n f ar mi ng .This is encompassed by the phenomenon of Big Data, massive volumes of data with a wide variety that can becaptured, analysed and used for decision-making. This review aims to gain insight into the state-of-the-art ofBig Data applications in Smart Farming and identify the related socio-economic challenges to be addressed. Fol-lowing a structured approach, a conceptual framework for analysis was developed that can also be used for futurestudies on this topic. The review shows that the scope of Big Data applications in Smart Farming goes beyondprimary production; it is influencing the entire food supply chain. Big data are being used to provide predictivei n si g ht s i n fa r m i ng op e r a ti on s, dr i ve r e a l -t i m e o pe ra ti on a l de ci s i on s, a nd re de s i gn bu si n e ss pr o ce ss e s f org am e - ch an gi n g b us i n e ss m o de l s. S e ve r al a ut h or s t he re f o re su g ge s t th at Bi g Da ta wi l l ca us e m a jo r s hi f ts inroles and power relations among different players in current food supply chain networks. The landscape of stake-holders exhibits an interesting game between powerful tech companies, venture capitalists and often small start-ups and new entrants. At the same time there are several public institutions that publish open data, under thecondition that the privacy of persons must be guaranteed. The future of Smart Farming may unravel in a contin-uum of two extreme scenarios: 1) closed, proprietary systems in which the farmer is part of a highly integratedfood supply chain or 2) open, collaborative systems in which the farmer and every other stakeholder in the chainnetwork is flexible in choosing business partners as well for the technology as for the food production side. Thefurther development of data and application infrastructures (platforms and standards) and their institutionalembedment will play a crucial role in the battle between these scenarios. From a socio-economic perspective,the authors propose to give research priority to organizational issues concerning governance issues and suitablebusiness models for data sharing in different supply chain scenarios.© 2017 T he Authors. Published by E lsevier L td. Th is is a n o pen access a rti cle u nder the C C B Y-NC-ND l icense(http://creativ ecommo ns.org/licenses/by-nc-nd/4.0/).Ke yw or ds :AgricultureDataInformation and communication technologyData infrastructureGovernanceBusiness modellingCo nt en ts1. Introduction.................................. ............................. 702. Methodology.................................. ............................. 713. Conceptualframework ............................. ............................. 713.1. Farmprocesses ............................. ............................. 723.2. Farmmanagement............................ ............................. 723.3. Datachain................................ ............................. 723.4. Networkmanagementorganization.................... ............................. 723.5. Networkmanagementtechnology..................... ............................. 724. Results .................................... ............................. 734.1. DriversforBigDatainSmartFarming................... ............................. 734.1.1. Pullfactors........................... ............................. 734.1.2. Pushfactors........................... ............................. 73Agricultural Systems 153 (2017) 69 80Corresponding author at: Wageningen University and Research, Hollandseweg 1, 6706KN Wageningen, The Netherlands.E-mail address: sjaak.wolfert@wur.nl (S. Wolfert).http://dx.doi.org/10.1016/j.agsy.2017.01.0230308-521X/© 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license ( http://creativecommons.org/licenses/by-nc-nd/4.0/ ).Contents lists available atScienceDirectAgricultural Systemsj ournal homepage: www.el sevier.c om /l ocate/ agsy4.2. Businessprocesses............................. ............................ 744.2.1. Farmprocesses........................... ............................ 744.2.2. Farmmanagement......................... ............................ 744.2.3. Datachain............................. ............................ 754.3. Stakeholdernetwork............................ ............................ 754.4. Networkmanagement ........................... ............................ 764.4.1. Organization............................ ............................ 764.4.2. Technology ............................ ............................ 774.5. Challenges................................. ............................ 775. Conclusionsandrecommendations......................... ............................ 77References....................................... ............................ 791. IntroductionAs smart machines and sensors crop up on farms and farm data growin quantity and scope, farming processes will become increasingly data-driven and data-enabled. Rapid developments in the Internet of Thingsa nd Clo ud C omp ut in g ar e p rop ellin g th e p hen om en on o f wh at is c al ledSmart Farming (Sundmaeker et al., 2016) . W hile Pr ec is ion Agr ic ul t ure isjust taking in-field va ria bility i n to a cc oun t, Sm ar t F arm in g go es be yon dthat by basing management tasks not only on location but also on data,enhanced by context- and situation awareness, triggered by real-timeevents ( Wolfert et al., 2014). Real-time assisting reconfi guration fea-tures are required to carry out agile actions, especially in cases of sud-de n ly ch a n ge d op er at ion a l c on di ti on s or ot h er ci rc um st an c es ( e. g.weather or disease alert). These features typically include intelligent as-s i s t an ce i n i m pl e m en t a ti o n , m ai n t e n an ce a n d u se o f t h e t ec hn ol o g y . Fig.1 su m ma ri z es th e c on c ep t of Sma rt Fa rm in g al on g th e m an a ge men tcycle as a cyber-physical system, which means that smart devices - con-nected to the Internet - are controlling the farm system. Smart devicesextend conventional tools (e.g. rain gauge, tractor, notebook) by addingautonomous context-awareness by all kind of sensors, built-in intelli-gence, capable to execute autonomous actions or doing this remotely.In this picture it is already suggested that robots can play an im portantrole in control, but it can be expected that the role of humans in analysisa nd pl an n in g is in c re asi n gl y a ssi st ed by ma c hi n es so th at th e c yb er -ph ys ic a l c yc le be c om es a lm os t a ut on om ou s. H um a ns wi ll a lwa ys bein volv ed in t he wh ol e p roc es s bu t in crea si n gly at a m uc h h igh er i n telli-gence level, leaving most operational activities to machines.Big Data technologies are playing an essential, reciprocal role in thisdevelopment: machines are equipped with all kind of sensors that mea-sure data in their environment that is used for the machines' behaviour.This varies from relatively simple feedback mechanisms (e.g. a thermo-stat regulating temperature) to deep learning algorithms (e.g. to imple-ment the right crop protection strategy). This is leveraged by combiningwith other, external Big Data sources such as weather or market data orbenchmarks with other farms. Due to rapid developments in this area, aunif ying defi ni tion of Big Dat a is diffic ult to give , but gen erall y it is aterm for data sets that are so large or complex that traditional data pro-cessing applications are inadequate (Wikipedia, 2016). Big data requiresa set of techniques and technologies with new forms of integration toreveal insights from datasets that are diverse, complex, and of a massivescale ( Hashem et al., 2015 ). Big Data represents the information assetscharacterized by such a high volume, velocity and variety to require spe-ci fic technology and analytical methods for its transformation into value( De Mau ro et a l., 201 6 ). Th e Dat a FA IRpo rt ini tia tive emph asiz es themore operational dimension of Big Data by providing the FAIR principlemeaning that data should be Findable, Accessible, Interoperable and Re-us able (Data FAIRport, 2014 ). This also implies the importance of meta-da ta i. e. data about the data (e.g. time, location, standards used, etc.).Both Big Data and Smart Farming are relatively new concepts, so it isex pec ted th at kno wle dge a bout the ir app lic ati ons an d the ir impl ic a-tions for research and development is not widely spread. Some authorsrefer to the advent of Big Data and related technology as another tech-nology hype that may fail to materialize, others consider Big Data appli-cations may have passed the pe ak of i nflated expectations in Gartner'sHype Cycle (Fenn and LeHong, 2011; Needle, 2015 ). This review aims toprovide insight into the state-of-the-art of Big Data applications in rela-tion to Smart Farming and to identify the most important research anddevelopment challenges to be addressed in the future. In reviewing thelit eratu re, a tte ntion is paid to bot h tec hni cal and so ci o-eco nom ic as-pe c ts . H ow ev er , t ech n ol og y is c ha n gi ng ra pi dl y in th is a re a a nd astate-of-the-art of that will probably be outdated soon after this paperis published. Therefore the analysis primarily focuses on the socio-eco-nomic impact Big Data will have on farm management and the wholenetwork around it because it is expected that this will have a longer-lastin g effect . From that persp ective the resear ch questio ns to be ad-dressed in this review are:1. What role does Big Data play in Smart Farming?2. What stakeholders are involved and how are they organized?3. Wh at a re th e ex p ec te d ch a n ge s th at a re c a us ed b y B ig D atadevelopments?4. Wha t c h al le n ge s nee d to be add re sse d in re la tio n to th e pr ev iou sques ti on s?The latter question can be considered as a research agenda for thefu tu re .To a nsw er the se qu es tio ns a nd to st ru ctu re the re vie w p roc es s, ac onc ep t ual fr a mew or k f or a na ly si s ha s be en de ve lop ed , wh ic h isexpected to be useful also for future analyses of development s in BigDa ta an d Sm ar t Far m in g. In th e re ma in de r of t hi s pap er th eFig. 1. The cyber-physical management cycle of Smart Farming enhanced by cloud-basedevent and data management (Wolfert et al., 2014).70 S. Wolfert et al. / Agricultural Systems 153 (2017) 6980meth odol ogy for revi ewin g the lit erat ure (Sect ion 2) an d the frame -work will be de sc ri be d (Sec ti o n 3). Then the main results from the anal-ysis will be presented in Se ct io n 4. Se ct io n 5concludes the review andprovides recommendations for further research and actions.2. MethodologyTo address the research questions as outlined in the In tr odu ction ,wesurveyed literature between January 2010 and March 2015. The choiceof the review period was a practical one and took into consideration thefact that Big Data is a rather recent phenomenon; it was not expectedth a t th er e wo ul d be an y re fe re n c e be f or e 20 10 . Be s id e t he pe ri od ofpublication, we used two inclusion criteria for the literature search: 1)full article publication; 2) relevance to the research question. Two ex-c lu si on c ri te ri a we re us ed : 1 ) a rt ic le s pu b li sh ed in la n gu a ge s ot h erthan English or Chinese; 2) articles focussing solely on technological de-si gn. Th e li ter a tu re su rv ey fo ll owe d a s ys tem ati c ap pro a ch . T hi s wa sdone in three steps. In thefirst step we searched two major bibliograph-ical databases, Web of Science and Scopus, using all combinations of twogrou ps of k eywor ds o f wh ic h th e first group addresses Big Data (i.e. BigDa ta, da ta- dr iv en in n ova ti on, dat a -d ri ven va lu e cre a tio n , in te rne t ofth in gs , Io T ) a nd th e s ec ond gr ou p re fe rs to fa rm in g ( i. e. ag ri c ul tu re ,fa rm in g, fo od , a gr i- fo od , pr ec is io n a gr ic ul tu re ) . T h e tw o da tab as eswere chosen because of their wide coverage of relevant literature andadvanced bibliometric features such as suggesting related literature orcitations. From these two databases 613 peer-reviewed articles were re-trieved. These were scanned for relevance by identifying passages thatwere addressing the research questions. In screening the literature, wefirst used the search function to locate the paragraphs containing thekey words and then read the text to see whether they can be relatedto the research questions. The screening was done by four researchers,with each of them judging about 150 articles and sharing their findingswith the others through the reference management software EndNoteX7. As a result, 20 were considered most relevant and 94 relevant. Theremaining articles were considered not relevant as they only tangential-ly touch upon Big Data or agriculture and therefore excluded from fur-th er re ad in g an d a na ly si s . W e f ou nd t he n um be r of re le va n t pe er -reviewed literature not very high which can be explained because BigData and Smart Farming are relatively new concepts. Especially the ap-pl ica tio ns a re ra pi dly ev olvi ng a nd ex pe cte d not to be tak en in to a c-c ou n t in pe er -r ev ie we d a rt ic le s wh ic h a re us ua lly la gg in g be h in d.The refo re we dec ided to also incl ude gre y lit eratu re in to our rev iew.For that purpose we have used Google Scholar and the search engineLexi sNexis for repor ts, magazine s, blog s, and other web -items in En-glish. This has res ulted in 3 reports , 225 magazine articles , 319 blog sand 19 items on twitter. Each of the 319 blogs was evaluated on rele-va nc e b as ed on it s ti t le an d se n te n c es co n ta in in g t he se a rch te rm s .Also possible duplications were removed. The result was a short list con-taining 29 blogs that were evaluated by further reading. As a result, 9blogs have been considered as presenting relevant information for ourframework. Each of the 225 magazine articles was similarly evaluatedon their relevance based on its title and sentences containing the searchterms. After removing duplicates, the result is a short list of 25 articles.We then read these 25 articles through for further evaluation. Conse-quently 9 articles have been considered as containing relevant informa-tion for further analysis.In the second step, we read the selected literature in detail to extractthe in fo r matio n rel ev ant to our r esea rch qu e sti on s. A ddit io nal li ter -ature that had not been identifi ed in thefi rst step was retrieved i nt h is s te p a s we l l i f t h ey w er e r e fe r re d to b y t h e m o s t re le va n tlit er-ature . Th is s n ow -b a l l a p pr oa c h h a s r es u l te d in 1 1 a d d it io n a l a rt ic le sand web-ite ms from which r elevant infor mation was e xtracted aswell. In the third step, th e extracted informatio n was analysed andsynthesized foll owing the con ceptual framework as described inSe ct io n 3.3. Conceptual frameworkFor this review a conceptual framework was developed to provide asystematic classi fication of issues and concepts for the analysis of BigData applications in Smart Farming from a socio-economic perspective.A major complexity of such applications is that they require collabora-tion between many different stakeholders having different roles in thedata value chain. For this reason, the fram ework draws upon literatureon c h ai n n et wo rk m an ag e men t a nd d ata -d ri ve n st ra te gi es . Ch ai nnetworks are considered to be composed of the actors which verticallyan d hor iz on tal ly wo rk to ge th er to a dd va lu e to c u st om er s(Christopher, 2005; Lazzarini et al., 2001; Omta et al., 2001 ). An impor-tant foundation of chain networks is the conceptv a lu e c ha in, which is asystem of interlinked processes, each adding value to the product of ser-vice (Porte r, 19 85). In big data applications, the value chain refers to thesequence of activities from data capture to decision making and datamarketing (Chen et al., 2014; Miller and Mork, 2013 ).Th e of te n -ci te d c on cep tu a l fr a me wor k of La mb er t a n d C oop er(2000)on network management comprises three closely interrelatedelements: the network structure, the business processes, and the man-agement components. The network structure consists of the memberfirms and the links between these firms. Business processes are the ac-tivities that produce a speci fic output of value to the customer. The man-ag e men t c om po n en ts a re th e ma n ag er ia l va ri a bl es by wh ic h th ebu si n es s pr oc es s es a re in te gr a te d an d ma n ag ed ac ro ss th e n et wo rk .The network management component is further divided into a technol-ogy and organization component.For our purpose the framework was tailored to networks for Big Dataapplications in Smart Farming as presented in Fig. 2.In this framework, the business processes (lower layer) focus on thegeneration and use of Big Data in the management of farming processes.For this reason, we subdivided this part into the data chain, the farmma n ag em en t a n d the fa rm pr oc es s es . Th e dat a ch a in in te ra c ts wi thfarm processes and farm management processes through various deci-sion making processes in which information plays an important role.Th e s tak eh ol de r n et wo rk ( m id dl e la ye r) c om pr is es a ll st ak eho ld er sth at ar e in vo lv ed in t hes e pr oc es s es , not onl y u se rs of Bi g D ata b utalso companies that are specialized in data management and regulatoryand policy actors. Finally, the network management layer typifies theorganizational and technological structures in the network that facili-tate coordination and management of the processes that are performedby the actors in the stakeholder network layer. The technology compo-nent of network management (upper layer) focuses on the informationinfrastructure that supports the data chain. The organizational compo-nent focuses on the governance and business model of the data chain.Fi n al ly , se ve ra l fa c to rs c a n be id ent ifi ed a s ke y d ri ve rs f or th eFig. 2. Conceptual framework for the literature analysis(Adapted from Lambert and Cooper (2000)).71 S. Wolfert et al. / Agricultural Systems 153 (2017) 6980development of Big Data in Smart Farming and as a result challenges canbe derived from this development.The n ext su bse ctio ns pr ovid e a more det ai led de scri pt ion of ea chsubcomponent of the business processes layer and network manage-ment layer of the framework.3 .1. Fa rm pr oc es sesA bus in es s p roc es s is a se t of log ic al ly re la te d ta sks pe rf or med toachieve a defined business outcome (Davenport and Short, 1990 ). Busi-ness processes can be subdivided into primary and supporting businessprocesses (Davenport, 1993; Porter, 1985 ). Primary Business Processesare those involved in the creation of the product, its marketing and de-live ry to t he bu ye r ( Por te r, 19 85). Supporting Business Processes facilitatethe development, deployment and maintenance of resources requiredin primary proce sses. The business processes of farming significant lydi f fe r be tw ee n di ff er ent t yp es of pr od u c ti on, e. g. li ve s to c k fa rm in g,arable farming and greenhouse cultivation. A common feature is thatagricultural production is depending on natural conditions, such as cli-mate (day length and temperature), soil, pests, diseases and weather(Nu th all, 20 11).3 .2. Fa rm ma na gemen tManagement or control processes ensure that the business processob jec tiv es a re a ch iev ed, eve n if dis tu rb anc es oc c ur. T he ba sic id ea ofcontrol is the introduction of a controller that measures system behav-iour and corrects if measurements are not compliant with system objec-ti ve s . Ba si c a ll y, th is im p lie s th a t t hey mu st ha ve a f ee dba c k lo op inwhich a norm, sensor, discriminator, decision maker, and effector arepr es en t (Beer, 1981; in 't Veld, 2002 ). As a consequence, the basic man-agement functions are ( Verd ouw e t a l. , 2 015)(seealso Fig. 1):Sensing and monitoring: measurement of the act ual per formance ofthe farm processes. This can be done manually by a human observeror automated by using sensing technologies such as sensors or satel-lites. In addition, external data can be acquired to complement directobservations.An al ys i s a nd de ci si o n ma k in g: c om pa re s mea su re men ts wi th th enor m s th a t sp eci f y th e de s ir ed pe rf or ma nc e ( s ys te m ob j ect ive sconcerning e.g. quantity, quality and lead time aspects), signals devi-ations and decides on the appropriate intervention to remove the sig-nalled disturbances.Intervention: plans and implements the chosen intervention to correctthe farm processes' performance.3.3. Data chainThe data chain refers to the sequence of activities from data captureto decision making and data marketing ( Chen et al., 2014; Miller andMork, 2013). It includes all activities that are needed to manage datafor farm management. Fig. 3 illu st ra tes th e m ain st eps i n th is c ha in .Being an integral part of business processes, the data chain consistsnecessarily of a technical layer that captures raw data and converts itin to in fo rma tion a nd a b us in es s l a yer th at ma kes de cisio ns an d d erive svalue from provided data services and business intelligence. The twola ye rs c an be in te rw ov en in ea ch st a ge a nd to ge th er the y f or m th eba si s of wh at h as c om e t o b e k now n as th e da ta va lu e c ha in(Du mbill , 20 14)(Ta ble 1).3.4. Network management organizationThe network management organization deals with the behaviour ofthe stakeholders and how it can be in fluenced to accomplish the busi-ness process objectives. For the uptake and further development of BigData applications, two interdependent aspects are considered relevant:governance and business model. Governance involves the formal andinformal arrangements that govern cooperation within the stakeholdernetwork. Important arrangements for the management of big data in-clude agreements on data availability, data quality, access to data, secu-rity, responsibility, liability, data ownership, privacy and distribution ofcosts. Three basi c forms of network governance can be distingu ished(Lazzarini et al., 2001): managerial discretion, standardization and mu-tual adjustment. These forms correspond with the three forms of net-wo rk gov er n a n ce pr es en te d by Pr ov a n a nd K en is (2 008 ) :leadorganization-governed network, network administrative organization,and shar ed par tic ipan t-g ove rned netw ork . Th e ch oic e of a par ticu la rnetwork governance structure aims at mitigating all forms of contractu-al hazards found between the different contracting parties in such a waythat transaction costs are minimized ( Wi llia ms on , 1 99 6). When study-ing hybri d forms of organiz ation su ch as supply chain networks, twomain dimensions should be identified: the allocation of decision rights,i.e., who has the authority to take strategic decisions within the supplychain network, and the inter-organizational mechanisms aim ing at re-wa rd in g de s ir ab le b eh av io ur a n d pr ev en ti n g u nd es ir a bl e be h a vi ou r(risk and rewarding mechanisms).De s pi te ag re em en t on th e im po rt a n c e of b us in es s m od el to anorganization's success, the concept is still fuzzy and vague, and thereis li tt le c ons en su s re ga rd in g it s c om p os it io n al fa c et s. Os te rw al d er(2004)defines business model as “… aconceptualtoolthatcontainsase t of el eme n ts a nd th ei r re la ti on sh ip s a n d a ll ows e xp res si ng acompany's logic of earning money. It is a description of the value a com-pany offers to one or several segments of customers and the architec-tu re of t he fi rm a nd it s n et wo rk of p ar tn er s fo r cre a ti n g, ma rk et in gand delivering this value and relationship capital, in order to generateprofitable and sustainable revenue streams. This definition re flects aso-calledfirm-centric view of business model. Another view on busi-ness model is the network-centric business model which builds uponva lue n et wo rk th eo ries ( Al-Debei and Avison, 2010). The value networkth eo ri es co n s id er b ot h fi nanc ia l a nd non - fi n an c ia l va lu e of b us in es str a ns ac ti on s an d ex c h an ge s. Bo th vi ew s a re re le va nt to t he n et wor kmanagement of Big Data applications.3.5. Network management technologyThe network management technology includes all computers, net-works, peripherals, systems software, application packages (applicationso ft wa re ), p ro ced ur es , t ech n ic al , in fo rm ati on an d c om m un ic ati onFig. 3. The data chain of Big Data applications, based on Chen et al. (2014).72 S. Wolfert et al. / Agricultural Systems 153 (2017) 6980standards (reference information models and coding and message stan-dards) etc., that are used and necessary for adequate data managementin the inter-organizational control of farming processes ( van der Vorstet al., 2005 ). Components to be mentioned here encompass:Data resources stored in shared databases and a shared understandingof its content (shared data model of the database).Information systems and services that allow us to use and maintainthese databases. An information system is used to process informationn ec es sa ry to pe rf or m us ef ul a c ti vi ti es u si n g ac ti vi ti es , fa ci li ti es ,methods and procedures.Th e wh ol e se t of f or ma li se d c od in g an d mes sa ge st an da rd s ( bo thte c hn ic al ly an d c on te n t- wi se ) wi th as so c ia t ed p ro ced ur es fo r u se ,connected to shared databases, which are necessary to allow seamlessand error-free automated communication between business partnersin a food supply chain network.The necessary technical infrastructure. None of the above can work ifwe don't have the connected set of computers (workstations of indi-vidual associates or people employed by or interested in the networkand the database, communication and application servers and all as-sociated peripherals) that will allow for its usage.In conclusion, this framework now provides a coherent set of ele-ments to describe and analyse the developments of Big Data in SmartFarming. The results are provided in the next section.4. Results4.1. Drivers for Big Data in Smart FarmingThere has been a signi ficant trend to consider the application of BigDa ta te c h ni qu es an d met h od s to ag ri c ul t ur e a s a ma jo r op p or tu n it yfor application of the technology stack, for investment and for the real-isation of additional value within the agri-food sector ( Noyes, 2014; Sunet al., 2013b; Yang, 2014 ). Big data applications in farming are not strict-ly about primary production, but play a major role in improving the ef-ficiency of the entire supply chain and alleviating food security concerns(Chen et al., 2014; Esmeijer et al., 2015; Gilpin, 2015a ). Currently, bigdata applications discussed in the literature are taking place primarilyin Europe and North America ( Faulkner and Cebul, 2014). Consideringthe growing attention and keen interest shown in the literature, howev-er, the number of applications is expected to grow rapidly in other coun-tr ies lik e C hi n a (Li et al ., 20 14 ; L iu et a l., 20 12). Big data is the focus of in-de p th , a dv an c ed , ga m e- c h an gi n g bus in es s an a ly ti cs , a t a s ca le a n dsp ee d th at th e ol d a pp ro a ch of c op yi n g an d c le an si n g al l of it in to adata warehouse is no longer appropriate ( Devlin, 2012). Opportunitiesfor Big Data applications in agriculture include benchmarking, sensorde p loy m en t an d a na ly ti c s, pr ed ic t iv e m od el li n g, an d us in g be t te rmodels to manage crop failure risk and to boost feed efficiency in live-stock production (Faulkner and Cebul, 2014; Lesser, 2014 ). In conclu-si on, Bi g D ata is to p rov id e p red ic ti ve in si ght s to f ut ur e ou tc om es offa rm in g ( pr ed ic ti ve yi el d m od el , pr ed ic ti ve f ee d in tak e mo de l, et c. ),drive real-time operational decisions, and reinvent business processesfo r fa st er , in n ov a ti ve a c ti on an d ga m e- c h an gi n g bu si n es s m od el s( De vlin, 201 2 ). Deci sio n-m aking in the futu re will be a c omple x mi xof human and computer factors (Anonymous, 2014b ). Big data is ex-pected to cause changes to both the scope and the organization of farm-in g ( Pop pe et a l. , 2 01 5). Whi le t her e ar e d ou bt s whe the r fa rm er s 'knowledge is about to be replaced by algorithms, Big Data applicationsar e li ke ly to c h an ge t he wa y fa rm s a re op er a te d a n d m an a ge d(Dr uc ke r, 2 014). Key areas of change are real-time forecasting, trackingof physical items, and reinventing business processes (Devlin, 2012 ).Wider uptake of Big Data is likely to change both farm structures andthe wider food chain in unexplored ways as what happened with thewi de r ado p ti on of tr a cto r a nd th e in t rod uc ti on of pe s ti ci d es in th e19 50s .As with many technological innovations changes by Big Data appli-cations in Smart Farming are driven by push-pull mechanisms. Pull, be-cause there is a need for new technology to achieve certain goals. Push,be c a us e n ew te c h n olo gy en ab le s p eo pl e or or ga n iz at ion s to ac h ie vehigher or new goals. This will be elaborated in the next subsections.4.1.1. Pull factorsFrom a business perspective, farmers are seeking ways to improveprofitability and efficiency by on the one hand looking for ways to re-duce their costs and on the other hand obtaining better prices for theirpr od u ct . T her ef or e th ey nee d to tak e b et te r, mo re op ti m al de c is io n sand improve management control. While in the past advisory serviceswere based on general knowledge that once was derived from researchexperiments, there is an increasing need for information and knowledgeth at is gen era ted o n -fa rm in its l oc al -s pe ci fic context. It is expected thatBig Data technologies help to achieve these goals in a better way ( Pop pee t a l . , 2 0 15 ; S o n ka , 2 01 5 ). A specific circumstance for farming is the in-flu en ce o f th e wea th er a nd es pe ci al ly i ts vola tility. L oc al-s pec ificweath-er a n d cl im ate da ta c an hel p de c is io n-m ak in g a lo t (L es s er , 2 014 ). Ageneral driver can be the relief of paper work because of all kind of reg-ulations in agri-food production ( Poppe et al., 2015 ).From a public perspective global food security is often mentioned asa main driver for further technological advancements (Gilpin, 2015b;L es se r, 2 014 ; Po pp e et al . , 2 015 ). Be s id es , c on su m er s a re be c om in gmore concerned about food safety and nutritional aspects of food relat-ed to he al th a nd well-b ein g ( Tong et al., 2015). In relation to that, Tonget a l. (2 015 ) men ti on th e n ee d fo r ea rl y wa rn in g sy st ems in st ea d ofmany ex-post analyses that are currently being done on historical data.4.1.2. Push factorsA ge n er al f ut ur e de ve lo pm ent is th e In te rn et of Th in gs ( Io T ) inwhich all kinds of devicessmart objects are connected and interactwith each other through local and global, often wireless network infra-structures (Porter and Heppelmann, 2014). Precision agriculture can beconsidered as an exponent of this development and is often mentionedas an important driver for Big Data ( Lesser, 2014; Poppe et al., 2015).This is expected to lead to radical changes in farm management becauseof access to explicit information and decision-making capabilities thatwe re pr ev io us ly not po ss ib le , ei th er te c h n ic a ll y or ec ono m ic a ll y(Sonka, 2014). As a consequence, there is a rise of many ag-tech compa-nies that pushes this data-driven development further ( Lesser, 2014).Wi re le ss da ta tr a n sf er te c h nol og y al s o pe rm it s fa rm er s to ac c es stheir individual data from anywhere whether they are at the farm-h ou se or me et in g wi th buy er s in C hi c a go ena bl in g th em to m ak eTable 1Key stages of the data chain on technical and business layer.Laye r of data chain Stages of a data chainRaw materia l Processing Transport MarketingTechnical Data generation and capture Data janitorial work, Data transformationData analyticsData transfer Data transferData analyticsBusiness Data discoveryData warehousingInterpreting data,Connecting data to decision(Obtaining business information and insight)Information share and data integration Data-driven services73 S. Wolfert et al. / Agricultural Systems 153 (2017) 6980infor med decision s abou t crop yield , harves ting, an d how best to getth ei r pro du ct to m ark et (Faulkner and Cebul, 2014).Table 2provides an overview and summarizes the push and pull fac-tors that drive the development of Big Data and Smart Farming.4 .2. B usi ness p roc esse s4.2.1. Farm processesAgricultural Big Data are known to be highly heterogeneous ( Ishii,2014; Li et al., 2014 ). The heterogeneity of data concerns for examplethe subject of the data collected (i.e., what is the data about) and theways in which data are generated. Data collected from thefield or thefarm include information on planting, spraying, materials, yields, in-sea-son imagery, soil types, weather, and other practices. There are in gen-eral three categories of data generation (Devlin, 2012; UNECE, 2013):( i) pr oc es s -m ed ia te d ( PM ) , (i i) m a ch in e- ge n er at ed (M G) an d ( ii i)human-sourced (HS).PM dat a , or th e tr adi tio n al bus in es s da ta, res ul t fr om ag ri cul tu ra lprocesses that record and monitor business events of interest, such aspu rch a si n g in pu ts , fe ed in g, s ee di n g, ap pl yi n g f er ti li zer , ta ki n g anorder, etc. PM data are usually highly structured and include transac-tions, reference tables and relationships, as well as the metadata thatdefine their context. Traditional business data are the vast majority ofwhat IT managed and processed, in both operational and business infor-mat ion syst ems , us ual ly st ruc tur ed and sto red in re lati ona l dat abas esystems.MG data are derived from the vast increasing number of sensors andsmart machines used to measure and record farming processes; this de-velopment is currently boosted by what is called the Internet of Things(IoT). MG data range from simple sensor records to complex computerlogs and are typically well-structured. As sensors proliferate and datavolumes grow, it is becoming an increasingly important component ofthe farming information stored and processed. Its well-structured na-ture is suitable for computer processing, but its size and speed is beyondtraditional approaches. For Smart Farming, the potential of unmannedaerial vehicles (UAVs) has been well-recognized ( Faulkner and Cebul,2014; Holmes, 2014). Drones with infrared cameras, GPS technology,are transforming agriculture with their support for better decision m ak-ing, risk management ( Anonymous, 2014c). In livestock farming, smartdairy farms are replacing labour with robots in activitie s lik e feedingcows, cleaning the barn, and milking the cows (Grobart, 2012). On ara-ble farms, precision technology is increasingly used for managing infor-ma ti on a bo u t ea c h p la n t in th e fi el d (Vo gt , 20 13) . Wi th th es e n ewtechnologies data is not in traditional tables only, but can also appearin ot her f orm ats lik e s ou nds or i ma ge s ( Sonka, 2015). In the meantimesevera l advanc ed data anal ysis techniqu es have been dev elope d tha ttr ig ge r th e us e of da ta in im ag es or ot h er fo rm a ts (L es se r, 2 014 ;Noy es, 20 14).HM data is the record of human experiences, previously recorded inbo ok s a n d wo rk s of a rt , an d la te r in ph ot og ra p hs , a ud io an d vi d eo .Human-sourced information is now almost entirely digitized and storedeverywhere from personal computers to social networks. HM data areusually loosely structured and often ungoverned. In the context of BigDa ta a nd Sm ar t Fa rm in g, hu ma n -s ou rc ed da ta h av e ra re ly be endiscussed except in relation to the marketing aspects (Verhoosel et al.,2016). Limited capacity with regard to the collection of relevant socialmedia data and semantic integration of these data from a diversity ofsources is considered to be a major challenge ( Bennett, 2015).Ta ble 3provides an overview of current Big Data applications in re-lation to different elements of Smart Farming in key farming sectors.From the business perspective, the main data products along the BigData value chain are (predictive) analytics that provide decision supportto business processes at various levels. The use or analysis of sensor dataor similar data must somehow fit into existing or reinvented businessprocesses. Integration of data from a variety of sources, both traditionaland new, with multiple tools, is the first pre requ isit e.4.2.2. Farm managementAs Big Data observe rs point out: big or small, Big Data is still data( De vl in , 2 012 ). It mu st be ma n ag ed a nd a na ly se d to ex tr a c t it s f ul lvalue. Developments in wireless networks, IoT, and cloud computingare essentially only means to obtain data and generate Big Data. The ul-timate use of Big Data is to obtain the information or intelligence em-bodied or enabled by Big Data. Agricultural Big Data will have no realvalue without Big Data analytics (Su n e t a l. , 20 13b). To obtain Big Dataanalytics, data from different sources need to be integrated into lagoonsof d ata. In this process, data quality issues are likely to arise due to er-rors and duplications in data. As shown in Fig. 4, a series of operationson the raw data may be necessary to ensure the quality of data.Since the advent of large-scale data collections or warehouses, theso-called data rich, information poor (DRIP) problems have been perva-sive. The DRIP conundrum has been mitigated by the Big Data approachwhi c h h a s u nl ea sh ed in f or ma ti on in a m an n er th a t c an su pp or t in -formed - yet, not necessarily defensible or valid - decisions or choices.Thus, by somewhat overcoming data quality issues with data quantity,dat a ac cess res tri ctio ns wit h on- deman d cl oud co mpu ting , caus ativ ean a ly si s wi th c or re la t iv e da ta a n al yt ic s, an d m od el -d ri ve n wi th ev i-dence-driven applications (Tien, 2013 ).Big data on its own can offer a-hainsights, but it can only reliablyde li ver l on g- te rm bu si ne ss a dva nt ag e w h en f ully in tegr at ed with tr ad i-tional data management and governance processes ( De vl in , 2 01 2). BigDa ta pr oc es s in g de pe nd s on t ra di t ion a l, p roc es s- m ed ia t ed d ata a n dmetadata to create the context and consistency needed for full, mean-ingful use. The results of Big Data processing must be fed back into tra-di ti on al bu si n es s p ro ces se s to en ab le c h a nge a nd ev ol u ti on of th ebu si ness .Table 2S um ma r y of p u sh a n d pu l l fa c to r s th a t d ri v e th e d ev e l op me n t of B i g Da ta an d S ma r tFarming.Push factors Pull factorsGeneral technologicaldevelop ments- Internet of Things and data-driventechnologies- Precision Agriculture- Rise of ag-tech companiesSophis ticated technology- Global Navigation Satellite Systems- Satellite imaging- Advanced (remote) sensing- Robots- Unmanned Aerial Vehicles (UAVs)Data generation and storage- Process-, machine- and human--generated- Interpretation of unstructured data- Advanced data

[返回]
上一篇:Automatic Target Recognition for Synthetic Aperture Radar Images Based on Super-Resolution Generative Adversarial Network and Deep Convolutional Neural Network
下一篇:机载雷达图像目标识别模型仿真研究