物联网平台和MES区别

国外技术牛人论述，物联网云服务器和MES系统的关系。论证一个观点 为什么物联网云计算平台不可能取代MES 生成执行系统Table of contentsExecutive SummaryThe lloT WorldWhat loT Platforms DoThe Critical Role(s) of MES for lloTMES History245677A Dozen Things MEs Delivers for loTWhy Some Want MES to Vanish (And Why It Still Wont) 10So the Thing ls, loT Needs Mese 2018 YNO ADVISORS CRITICAL MANUFACTURINGIIOT HAS A THING FOR MES- WHY IOT PLATFORMS WONT REPLACE MES FOR INDUSTRY 4.0The llot worldToday's technology holdsThe specific version is Industrial Internet Incorporating lloT into products beforeout hope that manufacturers of Things (lon). loT uses"machinethey are built can shift production tocan orchestrate the complexlearning and big data technologya distributed and autonamous modeworld of production andharnessing the sensor data, machineThe lloT can store production ordersto-machine(M2 M)communicationand specifications on the device, orsupply chain with moreand automation technologies that have product being built. This enables thespeed and reliability thanexisted in industrial settings for yearsshop floor marketplace, described inever before. Perhaps theaccording to Tech Targetthe paper Manufacturing Software forbiggest new technology toIndustry 4.0.: Embracing Change andLeveraging existing technologies is onedrive that is the internet ofDecentralization for Success. In additionof the core strengths of oT. Those whoThings (IoTto directing the product throughfeel loT is not new have a point; mostthe plant, it will also work throug!manufacturers have been using m2Mthe supply chain and into use at thesensors and automation for yearscustomer for post-sale servicesTo have full transformative effect, loTbuilds on not only the automation but Usually, the promises for improvedalso the information systems already in business outcomes can justify these lOTplace to drive new insights by applying investments easily. However, gainingsensors and analytics and connectingthe benefits is not necessarily easy. Thethe smart devicesorchestration challenges shift in thisworld of distributed intelligence andlloT can make machines, materials anddecision-makingproducts"smart"and connected. Theseare a foundation for Industry 4.0, Smart This is so new many practitioners andManutacturing and Digitalizationexperts are debating and learning abouthow to manage operations in this lloTlloT for equipment can have a quickenvironment. A common argumentreturn and long-term benefits. Relativelyweve heard is that mES will vanish andunsophisticated equipment may notsimply become a set of apps runningant adding traditional sensors andn an loT platform. We do not agrstandard interfaces. but the external llotand find that the original reasons whyapproach may be easy to justifyMeS has become an integrated suitesophisticated equipment that doesapplications are even stronger nowhave sensors and standard interfacellot offers a fast and affordable watest additional process measurementsding on the results of the test,either the setbecome ac 2018 YNO ADVISORS CRITICAL MANUFACTURINGIIOT HAS A THING FOR MES- WHY IOT PLATFORMS WONT REPLACE MES FOR INDUSTRY 4.0What lot platforms doAt a high level, an InternetIn short, the lot platform is theanother source of information that theof Things(oT) platformplumbing between the data collected loT platform collects along the cycle ofis the middleware of thefrom"things"or devices at the edge and bringing the detailed data to the cloudnew way of working. Mostthe user-facing applications, as Figure1 shows. This does not mean all llotAlone, loT does not offer many newloT platforms include aplatforms are basiepossibilities to manufacturerscommunication networkThat's becauseand software for monitoring,Some of the most sophisticated loTtroubleshooting andplatforms do have capabilities for dataIt cannot pull in all the corcollection and aggregation2. It is not designed for orchestrationadministrating the connectedof data, alerts and notificationsacross one or more facilIties:devices and managing3. The lot devices have sensors withintegration to enterprise systems, forms,the network and datasome processing and communicationdashboards. collaboration and datamanagement. Some loTanalytics. These capabilities sound verycapabilities, but are not responsivelatforms have software for similar to what mES does and someTo summarize, the problem with thistranslating or analyzing theWOUad argue that adding apps forloT platform centric approach it thatdata and support for appspecific types of MES functionality willis not equipped to drive toward theallow the lot platform to provide all of business obiective The obiective isdevelopmentthe capabilities mES doesresponsiveness-for the equipmentor product to change its behavior toHowever, loT platforms act as a masterlayer from the bottom-up, meaningoptimize business outcomes. This isDO IIOT PLATFORMSfrom the equipment interfaces andwhat mes doesAND APPS MEANloT to the layers above MES is justMANUFACTURERS NOLONGER NEED MES?WE DONT BELIEVE SO: IN FACTIndustrial Things:ApplicationsMachines, devicesTHE NEW MES PLAYS A CRITICALROLE IN IIOT SUCCESSPlatformgAn loT platform connects smart devices througha cloud-based networkplicat ons, users andother data sources. It is intelligent midd ewareIES is one of the applications that an loI platformpulls from and feedsOther Data sourcesc 2018 YNO ADVISORS CRITICAL MANUFACTURINGIIOT HAS A THING FOR MES- WHY IOT PLATFORMS WONT REPLACE MES FOR INDUSTRY 4.0In this way, llot boosts rather thanprocesses that determine whether aincreases, this must occur throughdestroys the need for MES. Our viewcompany makes a profit or not. Fromprocesses. For example, manufacturerscomes from both long experience and that vantage, MES orchestrates theachieve consistent quality ot productsdeep analysis of the new rea ity.lower level equipment and devices inthrough quality of processes and quickalignment with these value-addiintroduction of new products throughMES acts top-down, and is moreprocesses. Since the ultimate goal ofeffcient processes across departmentsstrategic. It comes from the shop-floorIndustry 4.0 is to guarantee productivityThe critical roles of mes for loTAs intelligence becomesOften, these decisions need to beManufacturing Execution Systemsdistributed across the llotmade quickly -but there is a danger of (MES), as shown in Figure 2, is thecations that runs plant andand its connected devices andmaking them exclusively at the edgeset of aBecause of the tight interdependencemultiplant operations from a consistentequipment, orchestration isof all activities across a plant or all theset of data, much as erp runs therequired not just to ensureplants in a company cr supply chain,offices. It serves operators, engineers,the data gets to the rightthere's a high risk of optimizing onesupervisors, schedulers, quality andplace, but to make sense of it. process or area of the operation inmaintenance personnel directlyMES ensures that lot data isa way that sub-optimizes the overallIt also delivers plant information inin a context that allows peoplebusiness outcomes. This is where MEs ways that other people and applicationsplays a role. It's where mEs has always across the enterprise and supply chainto make sound operationalplayed a role, in fact. The differencecan use itand business decisionsthat now, the environment is changingChange Managementalytics IntelligenceReports- Dash- Digital Twin -Mobile GUICore Manufacturing FunctionsData Collection Tracking -- --- TraceabilityOperations SupportQuality-Maintenance -- Shift Log-Costing-Order MgmtEnterprise IntegrationAutomation IntegratiERP- PLM- SCM-HCMControls Recips llotgMES does provide connection to devices and to otherapplications, but in the middle, it's a rich set of functionsMaster Data Managementc 2018 YNO ADVISORS CRITICAL MANUFACTURINGIIOT HAS A THING FOR MES- WHY IOT PLATFORMS WONT REPLACE MES FOR INDUSTRY 4.0MES HIStoryDecades ago, it was important toconsistent system for both operations of transactions in those systemsspecify "integrated MES"becauseand operations support roles such asYet it does much morepreviously, separate point applications engineering, quality and maintenance(or paper or spreadsheet-basedto share status-in context. This viewSpecifically talking about data, thesystems) provided various functionsis available in near real-time, fastISA-95 idea that each level going upFor example, data collection, routings,enough for operations decisions and would fulfill a specific role of datawork instructions, quality, scheduling.for plant status to inform enterpriseaggregation, calculation or samplingrecipe management maintenancedecisionsmaking sure that only the relevantwere all managed in disparate piecesinformation will be passed up is nowof software. This is still the case inBeyond showing status andobsolete. This paradigm has beenmany plants, though most companiesproviding visibility across thedisrupted by big data and machineknow it's not optimalplant, the MES has always guidedearning applied on big sets ofand informed the peoplgranular data. These algorithms willIntegrated MES became the normthere. It does provide a middleware get the most value if no informationin industries such as semiconductorbetween the ERP and plants lost in data aggregation orand electronics largely because itautomation, bridging between thesimplificationdelivers a plantwide view MES is the very different speed and granularityA Dozen Things mEs Delivers for lIoTThere are things that mes Here are a dozen specific things thatdoes that an loT platformmes delivers that are important toand apps cannot. In additionlloT success (listed below and shownthere are ways in which MESin Figure 3)provides the optimal meansto fully leverage the benefitsPeople-orientedSupport forMachine-orientedof iot. a core function of mesprocessesis to manage and controlStreamlinedmanuFacturing processesOrchestrationStandardizationInnovationIt is not by accident thatthe word process, so criticalCollaborationResponsivenessSmart Surrogatefor manufacturing, is sorarely used in the mes-freeOrganizationConnectionModelintheoriesdE万EnforcementContextMES must provide advanced capabil ities in atVisibilityleast these 12 critical areas for lot to succeedc 2018 YNO ADVISORS CRITICAL MANUFACTURINGIIOT HAS A THING FOR MES- WHY IOT PLATFORMS WONT REPLACE MES FOR INDUSTRY 4.01. CONTEXT.process or quality anomalies and finaloccur this is where the value of llottest parameters. it can also deliverwith MES appears in better productionloT sensor data and machine learninginformation to engineers, showingoutcomesalgorithms may be adequate to detectproduct and process as-designed vsanomalies and even predict someactual parameters tied to throughput5. ORGANIZATION:failures. However, without more contextyield and environmental metricsMES maps and stores informationthose analytics can reach only a fractionof their potential MES can add data3. SMART SURROGATE.on operations, both processes andabout the specific product, equipmentassets. The llo can operate a facilityand recipe, plus data from previousPart of what's exciting about loT isautonomously only when all of thatsteps and equipment maintenancethat it adds intelligence to the devicesinformation is known. Without a systemhistory. One tell ing example is thatHowever, in many cases, there's noin which to organize and leverage thatbusiness case for adding sensors todata, lloT-based operations wouldvision systems with machine learningcannot tell sheepdogs from mops, orproducts and equipment. In those cases, be nearly impossible to understandkittens from ice cream! (Figure 4)MES can act as proxy for those devices. For example, just sensor data and anThe mes knows what processing theloT platform cannot determine the2. CONNECTION:product needs and can communicateequipment state. Without that view,the next step needed on behalf ofthe plant cannot see overall equipmentIn addition to being the set ofthe product It can also broadcasteffectiveness(OEE). OEE is a signal foapplications that run the plant, MESavailability on behalf of the equipment where to focus both immediate changesis designed to integrate with otherto ensure that products get through the in automation settings as well as longapplications. It delivers context at theproduction process efficientlyterm continuous improvement effortsright level naturally, since it managesproduction processes end-to-end. In4, RESPONSIVENESSthis way, mes can act as plant liaison tothe smart supply chain. For example,As mentioned above, lloT typically is notMES can show show procurement andresponsive. So MES can trigger changessuppler quality teams material receiptin routing, equipment behavior, ortime, incoming quality test results,operator instructions when anomaliesSHEEPDOG OR MOP?Figure 4: Image recognition with machine learning algorithms struggle to distinguish betweenhese Sheepdogs and cleaning mops. While the technology is improving, it has a distance to goc 2018 YNO ADVISORS CRITICAL MANUFACTURINGIIOT HAS A THING FOR MES- WHY IOT PLATFORMS WONT REPLACE MES FOR INDUSTRY 4.06. STANDARDIZATIONoperate as theyre supposed to, andthat production value is maintainedMES stores and allows easy andincreasedWhat is the New mesconsistent roll out of workflows orand how is it different?processes. Whether in a single plant10. VISIBILITYor many, this engineering time saverModern MES shows production statusThe plant modelwill be increasingly essentialin real-time across all assets. Mappingflexible not fixedOnce the local machine learning kicks in,utting it into the context of what worksthe in-context data onto the plantIt includes locationfor the plant, supply chain and businessmodel enables instant visibility In theprocessing forwill allow companies to confidentlylloT-enabled world, rather than knowingwhere a product is based on start timemobile interfaces andleverage best practices even in thedynamic lloT environmentand pre-set routing, the plant flooraugmented realitymarketplace will enable optimal paths7. COLLABORATION:given the current state of each producAdvanced analysis in theand piece of equipment. Nothing is pre-cloud to feed plant contextHaving a common information systemset. So, gaining real-time visibility acrossto other big data systemsacross plant operations (in one orthe entire facility -or set of facilitiesmultiple facilities) delivers consistentwill be more important than ever.Binding between assetscurrent data, allowing all operationsand processes for contextgroups to work together and use each 11 STREAMLINED INNOVATIONther's data with minimal friction thisresolutionis a major benefit of MES compared toOne of the maior benefits of Industr4.0 and loT-enabled production isDynamic execution offunction-specific appsmaking personalized or one-off product processes and status in8. MODELINGquickly and cost-effectively. The onlyshop floor marketplaceway to do that is to have MES showingMany companies implementing MESall of the context to enable rapid rampService-oriented, modularfind benefits before the software is inup and improvement in products andvisibility across operationsfull use. That's because mES is basedprocesseson a model of the plant(s), or a digitaltwin of the process. In the lloT world,12. ORCHESTRATION:the model is the framework on whichall of the dynamic activities betweenWhile an loI platform can keep datadevices can occur. When this model ismoving between devices, it cannotconnected into metrics about supplyoptimize plant operations outcomeschain and business results companiesThe 11 things meS does listed aboveare part of its ability to orchestratecan ensure their oT efforts optimizeboth plant and business resultsproduction operations. Separate appseven connected to a powerful llo9. ENFORCEMENTplatform cannot do that. Just as theorchestra has leads for every instrument,Manufacturing Execution Systemsthey still rely on the conductor to ensuremanage and enforce every aspect ofmaximum impact of the overall piece omanufacturing workflows, includingmusic or concertequipment, materials, procedures andautomatic or manrocesses. TheseSome of these things any MES wouldprocesses can be engineering relatedprovide; others require the new mES,(equipment, product or process), quality, described in the sidebar box and thegistics, etc. MES sets and manages the paper The New MES: Backbone ofpolicies and rules required to make sure Industry 4.0hat all entities, human or non-humanc 2018 YNO ADVISORS CRITICAL MANUFACTURINGlIOT HAS A THING FOR MES -WHY IOT PLATFORMS WONT REPLACE MES FOR INDUSTRY 4.0Why Some Want MES to Vanish(And Why It still Won'tLARGE SOLUTION PROVIDERSWhat these big players needed wasable to derive intelligence or insightsAND“ EXPERTS"a disruptive technology that wouldfrom this data. This is not dependentClearly, MES delivers big benefitsdissolve the segment differenceson the proliferation of loT. Yet IoT hasto lloT. So why have some solutionallowing the same technical solutions the potential to take it to a completelyproviders and industry analysts beento operate on all segments. Ideallynew level, given the amount of datainsistently conveying the message that this new technology would alsocollected (or the cost of generatingwith the advent of IIoT together withike deployment easy so that thesuch data, which will reach incrediblyloT platforms, machine learning andcustomers or a network of smallerlow levels)big data analytics, MES is no longersystem integrators could do it.MANUFACTURERSrequired?Apparently, they think they found it inSome manufacturers also echo theseWe can't say for sure, but perhapsIoT. The message is quite appealingMES-free strategies Here, what's atit's what they hope will be true. MESIoT devices(which will continue tostake is the prospect of less effortas always been difficult for largerbecome both more powerful andand faster returnenterprise application providers foraffordable) are connected to IOTtwo main reasons: segment specificity platforms that manage and controlThese companies are hoping theyand deployment effortssuch devices in turn the collectedcan bypass or circumvent all thedata points are stored in cloud-baseddifficulties that a standard maturity1. Sub-Segments: Large softwaresolutions There intelligent algorithmsprocess takes to get a company fromproviders find the manufacturinganalyze them and derive diagnostics.the least mature level to the mostapplications sector is too small topredicting the behavior of theThe thought here is: why should webe interesting when it's dividedunderlying processes and prescriptivehave to go through all the hurdlesinto meaningful groups or industry actions to improve their performanceof implementing prcgesegments. Discrete, batch andthem under control through peoplecontinuous have different needsIt sounds as if in this environmenttechnology and procedures and laterand specific industries within eachthere's no space (or need) for MESimprove it and supplement it withof those modes do alsosystems, whose main function wasadditional data? This lengthy processto collect and store data for decisiontakes time, effort investment and2. Deployments: Even with segmentmaking or for fulfilling qualitypersistence. Can't we do all this byspecific software, every company,requirementsimplementing Industry 4.0 and IoTplant or production area mighttechnology stack that does all ofAnd there's an additional angle: theneed configuration or customizationit at once?data collected is the new gold. Theto support best practices. This takessoftware players who control the data The problem is that companies cannottime and expertise to understandcan also offer additional value-addedskip or shortcut maturity phasesrequirements, design, implement,services on their platforms (at a goodWhatever the technology, no mattertest and deploy.selling price). Although Customers own how powerful the sensors, analyticsthe data, they will pay more and more and machine learning are, theres nofor solutions from providers that aremagic wand that will do the work.g 2018 YNO ADVISORS CRITICAL MANUFACTURING