字幕列表 影片播放 列印英文字幕 [music] Hi, I'm Michael Curry Vice President of product management for application and integration middleware inside of IBM Today I'm going to talk to you a little bit about the Internet of Things The Internet of Things is about weaving together the physical and digital worlds to deliver new value for different types of customer scenarios The demonstration I'm going to show you today focuses on a piece of physical equipment in an asset-intensive industry I'm going to show you how we're going to connect that to the Internet, to the Cloud, and actually begin to collect data from that and be able to react to changes in that physical equipment in a way that helps improve my business results So let's get into it In this case we will be using a pump I'm going to connect the pump into the Cloud and begin collecting data from the pump I'll be using IBM Bluemix our Cloud platform for rapid application development and deployment Since the Internet of Things always starts with a thing that is connected I'm going to be starting with this demonstration board which is a TI Texas Instruments sensor tag This sensor tag is connected into this BeagleBone board through bluetooth which is then connected through wifi into the Internet and eventually into the Cloud This is going to represent my pump So first I want to register this device into Bluemix and begin collecting data from it and what you can see is I'm viewing a live datafeed from the sensors You can see as a shake the device that the data is changing in real time The red line represents the vibrations but simply seeing the data is not enough developers want to build applications around the data and logic around the data so for that I'm going to use Bluemix In Bluemix we've created an Internet of Things boilerplate that deploys the services I need to connect, store, and build logic around the Internet of Things Inside Bluemix I have access to all the other services within the ecosystem including mobile services, Big Data services, Watson services, and third-party services I can use our DevOps tools to quickly build, test and deploy services, and embedded device code that leverage the power of Bluemix As I mentioned at the beginning this board represents the pump I want to set up logic to detect when an unusual vibration pattern occurs in the pump so I'll use my Bluemix design tool to define this logic This flow simply collects data from the pump, runs a predictive maintenance and quality service against the data which uses big data analytics to predict failures and then when a failure condition is sensed it will trigger a work order in our asset management service to dispatch a repair technician with the right skills, tools and parts before the failure happens It will also send a text message through Bluemix Twilio service to my phone So I'll simply link my predictive maintenance application into the flow and deploy it and now I'll be using predictive analytics against the realtime data stream to recognize patterns that would indicate the potential pending failure In this case we're only looking at three data streams off a single device to predict health but in a live scenario we could be monitoring over a hundred simultaneous sensor readings per device with thousands of devices PMQ, which is our predictive maintenance and quality service is using advanced predictive analytic models to determine the likelihood of failure based on historical and realtime data You can see now in my realtime datastream that when I simulate the vibration the predictive analytics recognize that the vibration exceeds the acceptable thresholds and it's making a recommendation on what I should do In this case is telling me to inspect and replace the impeller bearing It's also generating a work order in our asset management system that will dispatch a repair technician to inspect the pump and of course I just received the text on my phone So if I go into my asset management screens I can see the work order that was just generated for this pump My Asset Management service acts as a system of record for assets tracking every asset in my inventory By connecting it into the Internet of Things I can associate realtime sensor events to my assets to manage them more efficiently The power of the Asset Management service is that the technician has all the information they need to fix the problem at their fingertips including detailed data from the sensors, repair history, repair procedures and information on spare parts and special tools that are needed for the repair for this special part What I showed you was a complete loop from connecting a device, to building an application around it in Bluemix, to harnessing the power of our big data analytics, to feeding into operational systems a record to dispatch a repair technician and address a real business challenge for asset-intensive industries The power of IBM in the Internet of Things comes when you combine our strengths in Cloud, data, and engagement to deliver new types of applications very quickly across all different industries You can use our extensive set of tools within Bluemix to be able to build and prototype new applications and test huge populations of connected devices without ever having to connect anything You can use our big data analytics to try new types of analytic models, predictive models, physical analytics, and even different types of analytics around data streams You can use our cognitive tools like Watson to be able to build entirely new types of applications very quickly that take advantage of cognitive computing and artificial intelligence And you can tap into our vast ecosystems of partners and customers that are delivering new innovations into the market to create new applications and build new partnerships around your data Only IBM can combine these capabilities together in a composable framework in a scalable way that can deliver upon enterprise goals across all industries and across all geographies Thank you for your time today and we look forward to working with you [music]
B1 中級 物聯網--邁克爾-庫裡 (Internet of Things - Michael Curry) 80 7 richardwang 發佈於 2021 年 01 月 14 日 更多分享 分享 收藏 回報 影片單字