by Gary Mintchell | Jul 30, 2019 | Asset Performance Management, Manufacturing IT
I first met Quuppa and saw a demo of its real time locating system at the 2018 Hannover Messe. I have written about it here and here. The company has developed an interesting technology and application.
They wrote about industrial applications picking up, so I asked for an example. Below is a story about defining a problem, sourcing a solution, and then implementing it.
NGK Ceramics is a global specialist in the manufacturing of ceramic substrates used in catalytic converter applications for automotive, truck and off-road vehicles. The US manufacturing facility, located in Mooresville, North Carolina, covers more than 500k square feet with 365 days a year, twenty-four hours a day operations.
The facility was initially designed in 1988 to serve a limited geographical area in the US. However, with the business growing faster than expected and more areas being served by the same production plant, NGK faced a major challenge: how to grow the capacity of the North Carolina industrial plant. Efficiency was clearly the answer. As a first step, ASRS (Automated Storage and Retrieval Systems), together with AGV’s (Automated Guided Vehicles), were introduced to move pallets and materials in the shop floor without human intervention.
Even if this mitigated the problem, it was still not enough to manage high, yet variable, production demands in the long run. As a result, during production peaks, the pallets transporting both raw materials and semi-finished goods were temporarily stored all around the shop floor according to specific procedures. While this addressed the problem of lack of (ASRS) storage space, it introduced a significant new one, the additional time spent finding and moving pallets from one production phase to the next. At least two workers per shift were assigned to this task: just searching for and moving pallets.
In addition to this, at least once a year a complete plant inventory is required to verify all materials stored in the facility, but not yet shipped or sold. During this activity, the entire plant was surveyed, and all pallets were identified and verified against the data registered in the internal ERP system. This activity could take up to one week, with the slow down (if not interruption) of the production activities. Inevitably, any items lost or duplicated created an impact on the bottom line.
To deal with these issues, in 2017 NGK Ceramics decided to explore how solutions based on a Real Time Locating System (RTLS) could help by providing a Digital Twin of the manufacturing plant: the location of every pallet would be tracked continuously and that data would be synchronized with NGK’s MRP systems. This tracking of pallets provides a real-time view of where they are located in the industrial plant, with a number of supporting services to easily and rapidly search them and manage the production cycle.
TRACKING SOLUTION: REQUIREMENTS NGK Ceramics decided to evaluate a number of different scenarios for implementing a RTLS to track the progress of material and semi-finished goods throughout the flow of its manufacturing process. The key requirements to be addressed by the solution were:
• Configurable tracking accuracy: since the industrial plant covers a large area, with different uses of the spaces within the plant (production area vs. stocking areas vs. corridors), the ability to vary the location accuracy of asset tracking was important. In some areas, where the density of pallets is typically high (such as the warehouse) sub-meter accuracy is required in order to easily locate a specific pallet among the many stocked there. On the other hand, a 10 metre accuracy is sufficient in corridors or transit zones, where it is sufficient to track the presence of the pallet in the zone;
• Infrastructure cost: as NGK Ceramics facility is rather large, the number of RTLS antennas required to achieve the desired accuracy was clearly an important variable of the solution to be adopted. This impacted both the cost of the infrastructure as well as the costs related to the cabling (e.g., connectivity and power). Another factor was the cost of the tags to be attached to the pallets. This extended beyond the capital cost to also include the cost of replacing the batteries in the tags.
• Asset search and location functionality : NGK wanted this Digital Twin to be used in a variety of ways, from centralized systems to hand-held devices using a Google maps style red dot metaphor, so how the system was able to process the information and extract actionable knowledge for the final user (the worker in the shop floor) was important. This required addressing issues related to the usability and ergonomics of the system, Machine-2-Machine (M2M) application integration, while delivering on its intended use and the need to facilitate the searching and location of assets.
• Maturity of the solution: an enterprise-ready solution was requested. This refers to the support for active monitoring services of both the platform and the RTLS infrastructure. Any device or software component deployed in the facility needed to be monitored, with notifications sent in case of anomalies in the system. This includes the battery status of the devices/tags used for tracking the pallets.
DIGITISING THE PRODUCTION PROCESS NGK retained the services of Statler Consulting a specialist in the area of beacons and RTLS technologies, and issued a Request for Proposal (RFP) for a solution able to track in real-time the assets in their facility, and to deliver the necessary supporting services for the optimisation and real-time control of their production process. Among the many solutions proposed, ThinkIN was chosen as it proved to be the best match to the requirements identified by NGK. ThinkIN is an innovative IoT platform for real-time tracking, monitoring and control of assets and workforce in industrial environments.
ThinkIN technology is based on Quuppa4 RTLS for the high precision location of assets in the shop floor. Quuppa utilizes a unique combination of Bluetooth Low Energy (BLE) and the Angle of Arrival (AoA) methodology, as well as advanced location algorithms that have been developed over the course of more than 15 years, to calculate highly accurate, real-time indoor positioning, even in the most demanding environments, including inside warehouses and manufacturing facilities. The low-power system is a reliable, highly-customizable, scalable and costefficient solution for providing an accurate “dot on the map.”
ThinkIN platform provides a comprehensive set of services ranging from real-time support (e.g, asset search and location, alerts and geo-fencing, etc.), to Industrial IoT analytics. It also includes a number of tools to support the active monitoring of the infrastructure (both hardware and software) and a comprehensive set of user interfaces to explore the data collected and used to locate assets in real-time in the shop floor. In terms of tracking technology, Quuppa RTLS provided an optimal trade-off in terms of location accuracy, number of antennas required to cover the NGK facility and maturity of technology.
Overall 95 antennas are used to cover the complete NGK facility, with a location accuracy of approximately one meter in the areas of interest and approximately 5 meters in other areas. Different tag form factors were evaluated. Eventually, a custom Bluetooth Low Energy tag with a slim badge form factor was designed and manufactured in order to optimally align with NGK’s existing manufacturing process. The tag ensures 4+ years of life without battery replacement. Pallets, carrying products or semi-finished goods, are identified by means of their Product Travel Ticket (PTT), which includes all the necessary information about the kind of product manufactured, together with information on production stage ( e.g. forming line, firing in kilns, etc.). At the very beginning of the production process, an RTLS TAG is associated with the pallet Travel Ticket through a mobile application running on a scanner.
The application allows the scanning of both the QR code present on the PTT and the QR code on the TAG. This association creates a Digital Twin of the pallet, which is now tracked in real-time throughout its manufacturing process. The pallets can now be easily located through the ThinkIN mobile service. Additionally, plant-level views allow staff to monitor the status of the pallets across the entire facility, maintaining an always up-to-date inventory of all pallets stocked or moving in the facility.
Starting from ThinkIN open APIs, a dedicated mobile interface was created for an optimal utilisation of data over the shop-floor and to facilitate the work of employees in the search and location of pallets with a specific Travel Ticket. Figure 3: Tracking of assets in NGK facility Additional services delivered through the ThinkIN platform enable the quality control of pallets depending on their production stage, with alerts being triggered if the pallet moves into areas not allowed. To prevent this, a specific geo-localised workflow is imposed on the travel path of pallets depending on their production process. Warnings are raised when the specific workflow is not adhered to.
LOOKING AHEAD
The project started in 2017 with an initial pilot phase, and is now scaling up to the entire production plant with a possible extension in the coming years to other NGK manufacturing sites. NGK is planning to obtain a return on their investment in a 2 year time frame. Today we are in year two and ThinkIN solution is integrated with the production control system adding value to the manufacturing process by making the pallet searching process more effective.
ThinkIN’s platform has allowed NGK to digitize the shop floor by recreating the plant on screens accessible to all workers. Thanks to the data collected by tags and devices, workers can use the interface to find pallets around the manufacturing plant based on the information of the goods transported by the pallets, such as product type, bench number, kill cycle, and other key criteria for the production routing.
The efficiency of the shop floor was significantly increased thanks to ThinkIN for Industry. In the first year, NGK Ceramics reduced the costs of the wasted time searching for pallets and of the time spent doing the annual inventory. Thanks to the new solution, the inventory is constantly up-to-date. Moreover, the accuracy in tracking reduced the risk of accidents caused by the movement of pallets with forklifts in the shop floor searching for the needed pallet. ThinkIN for Industry, therefore, is a location intelligence technology that by capturing data from the shop floor in a digital platform offers the chance to automate the real world in new ways that can enhance and optimise workflows in the shop floor.
by Gary Mintchell | Jun 17, 2019 | Manufacturing IT, Software
Digitalization requires digital data, which in turn requires a place to robustly store that data. This place is most often the cloud these days. OSIsoft PI System must be the most widely used industrial database. The company has released OSIsoft Cloud Services—a cloud-native, real-time data management system for unifying and augmenting critical operations data from across an organization to accelerate industrial analytics, data science projects, data sharing, and other digital transformation initiatives.
OCS highlights and capabilities:
- Data sharing – partner companies can access a shared data stream to remotely monitor technology
- Functionality – seamless crossover between the PI System and OCS to compare facilities, perform root cause analysis and run hypotheticals
- Scalability – tests proved OCS can simultaneously manage over two billion data streams, and safely share information with partners
- Petuum uses OCS to stream historical data and live data on production, temperature and variability to its AI platform to assist Cemex, a global cement manufacturer, improve yield and energy to 7% from 2%.
- DERNetSoft uses OCS to aggregate data in one place, allowing users to access useful analytics for ways to reduce power and save money.
- Pharma companies will use OCS to give a regulator access to anonymized drug testing or production, without risk of unauthorized users in the manufacturing networks.
With OCS, an engineer at a chemical producer, for example, could combine maintenance and energy data from multiple facilities into a live superset of information to boost production in real-time while planning analysts could merge several years’ worth of output and yield data to create a ‘perfect plant’ model for capital forecasts.
OCS can also be leveraged by software developers and system integrators to build new applications and services or to link remote assets.
“OSIsoft Cloud Services is a fundamental part of our mission to help people get the most out of the data that is at the foundation of their business. We want their cost of curiosity to be as close to zero as possible,” said Gregg Le Blanc, Vice President of Product at OSIsoft. “OCS is designed to complement the PI System by giving customers a way to uncover new operational insights and use their data to solve new problems that would have been impractical or impossible before.”
The Data Dilemma
Critical operations data—i.e. data generated by production lines, safety equipment, grids, and other systems essential to a company’s survival—is part of one of the fastest growing segments in the data universe. IDC and Seagate estimate in “Data Age 2025: The Evolution of Data to Life Critical” that “hypercritical” data for applications such as distributed control systems is growing by 54% a year and will constitute 10% of all data by 2025 while real-time data will nearly double to more than 25% of all data.
Critical operations data, however, can be extremely difficult to manage or use.
Data scientists spend 50 percent or more of their time curating large data sets instead of conducting analytics. IT teams get bogged down in managing VPNs for third parties or writing code for basic administrative tasks. Data becomes inaccessible and locked in silos. Over 1,000 utilities, 80% of the largest oil and gas companies, and 65% of the Fortune 500 industrial companies already use the PI System to harness critical operations data, turning it into an asset for improving productivity, saving money, and developing new services.
Natively compatible with the PI System, OCS extends the range of possible applications and use cases of OSIsoft’s data infrastructure while eliminating the challenges of capturing, managing, enhancing, and delivering operations data across an organization. Within a few hours, thousands of data streams containing years of historical data can be transferred to OCS, allowing customers to explore, experiment, and share large data sets the same day.
Two Billion Data Streams
The core of OCS is a highly scalable sequential data store optimized for time series data, depth measurements, temperature readings, and similar data. OSIsoft has also embedded numerous usability features for connecting devices, managing users, searching, transferring data from the PI System to OCS, and other functions. OCS can also accept data from devices outside of traditional control networks or other sources.
“The scale and scope of data that will be generated over the coming decades is unprecedented, but our mission remains the same,” said Dr. J. Patrick Kennedy, CEO and Founder of OSIsoft. “OSIsoft Cloud Services represent the latest step in a nearly 40 year journey and there’s more to come.”
To test the scalability and stability of OCS, OSIsoft created a deployment that contained the equivalent of the data generated by all of the smart meters in the U.S. over the last two years, or two billion data streams (100 million meters with 20 data streams each). OCS successfully stored up to 1.2 billion data points per hour and was managing all two billion streams simultaneously within 48 hours.
PaaS for OSIsoft Marketplace Partners
Software developers are already creating services based around OCS. DERNetSoft is creating a secure marketplace for sharing utility and electric power data to improve energy forecasts and peak shaving strategies. Meanwhile, others are collaborating with customers on efforts to bolster well integrity at oil drilling sites, pinpoint tank leakage, predict maintenance problems, and reduce energy consumption with OCS. OSIsoft partners developing OCS services include Petuum, Seeq, Toumetis, Transpara, Aperio, and TrendMiner. These services will be available from OSIsoft marketplace as they are released.
“Digital transformation requires the ability to compare data and outcomes across multiple plants and data sources,” says Michael Risse, VP/CMO at Seeq. “OCS is a unified solution for process manufacturing customers to enable this type of analysis, generating predictive insights on thousands of assets across company operations to improve production outcomes.”
Pricing and Availability
OCS is a subscription service currently available to customers and partners for use in facilities in North America. OCS will be extended to Europe and to other regions in the near future.
Pricing is based on the average number of data streams accessed, rather than the unique data streams stored, giving customers the freedom to experiment more freely with their data without incurring added costs..
by Gary Mintchell | Nov 28, 2018 | Internet of Things, Manufacturing IT, Operations Management, Technology
Hewlett Packard Enterprise (HPE) announced new HPE Edgeline Converged Edge System solutions that speed the deployment and simplify the management of edge applications, enabling customers to act on the vast amounts of data generated by machines, assets and sensors from edge to cloud.
I think this is another significant advance reflecting the utility of enterprise compute capability brought ever closer to the plant itself. If you are looking to be disruptive in your industry or are on a corporate engineering staff looking for OT alternatives, I’d suggest taking a long look at these technologies and then letting your imagination do its work.
The new solutions include:
- HPE Edgeline OT Link Platform, an open platform that automates the interplay between diverse operational technologies (OT) and standard IT-based applications at the edge to enable intelligent and autonomous decision making;
- HPE Edgeline systems management, the industry’s first systems management solutions designed specifically for the edge to ensure enterprise-grade reliability, connectivity and security;
- HPE Edgeline EL300 Converged Edge System featuring OT link and HPE Edgeline systems management, providing superior resilience against harsh edge environments for a broad range of industrial deployments; and
- HPE Edgeline Field Application Engineering Services are available from HPE Pointnext to help customers plan, build, and customize OT link-based Internet of Things (IoT) and cyber-physical systems.
To turn edge data into insight for real-time action, it must be processed close to its source to avoid the latency, bandwidth, and cost issues of sending the data to a remote data center. However, this opportunity comes with a set of unique challenges, including management of remote infrastructure, and the necessity to seamlessly connect sensors and industrial assets with IT applications at the edge.
“Deploying IoT, edge, and cyber-physical systems is a challenge requiring a fresh look at uniting the physical and digital worlds,” said Dr. Tom Bradicich, Vice President and General Manager, Converged Servers, Edge and IoT Systems, HPE. “With today’s announcements, we enable our customers to accelerate the delivery of applications that capitalize on edge data, safeguarded by enterprise-class management. And we lay the groundwork for a new ecosystem of intelligent edge solutions to drive innovation and growth across industries.”
Simplifying deployment of edge-to-cloud IoT and cyber-physical systems
Today, setting up an IoT or cyber-physical system is a laborious undertaking. It requires custom coding to orchestrate OT networks, control systems, and data flows with drivers, middleware, and applications running on IT systems. HPE Edgeline OT Link Platform is an open platform that significantly simplifies this process, reducing cost and time to market.
The solution includes:
HPE Edgeline OT Link Platform software, an open workflow engine and application catalogue, allowing customers to orchestrate components, data, and applications via a graphical drag-and-drop user interface. The HPE Edgeline OT Link Platform integrates an ecosystem of third-party applications running from edge to cloud – including AWS, Google, Microsoft, SAP, PTC, GE, and more – to make insights from the edge available across the enterprise and supply chain.
HPE Edgeline OT Link certified modules, HPE-developed adapters that connect to a broad range of OT systems, enabling bi-directional, time-sensitive, and deterministic control and communication, including high-speed digital input/output, CAN bus, Modbus, or Profinet. APIs and SDKs for these adapters are made available to the industry to facilitate third-party designs of OT link modules. OT link will also integrate FPGA modules to give customers maximal flexibility to connect to any industrial input/output device.
Enterprise-grade manageability and security at the edge
HPE also announced the industry’s first systems management solutions specifically designed to simplify the provisioning and management of edge infrastructure and applications, providing enterprise-grade manageability and security for remote systems with limited connectivity and IT expertise.
HPE Edgeline Integrated System Manager is embedded into HPE Edgeline Converged Edge Systems and features one-click provisioning, ongoing system health management, remote updates, and management even with intermittent wired and wireless connections. It also supports advanced security functions like preventing system boot file changes and remote system disablement during a security event. HPE Edgeline Infrastructure Manager software can remotely manage thousands of Edgeline Converged Edge Systems.
The HPE Edgeline Workload Orchestrator hosts a central repository for containerized analytics, AI, business, and IoT applications that can be pushed to HPE Edgeline Converged Edge Systems at the edge
Unparalleled convergence of OT and IT
The HPE Edgeline EL300 is a fan-less, low-energy system equipped with Intel Core i5 processors, up to 32GB of memory and 3TB of storage. It will also support Intel Movidius Myriad X vision processing units to enable video analytics and AI inference at the edge. The HPE Edgeline EL300 provides enhanced resiliency against shock, vibration, humidity, and dust, including IP50 and MIL-SPEC certifications, and can operate from -30 to +70 degrees Celsius. These features make the HPE Edgeline EL300 suitable to be deployed as an embedded system – for example, in production machines or in building infrastructure.
Expertise to accelerate deployment and create competitive advantage
To support these new offerings, HPE Pointnext, the services organization of Hewlett Packard Enterprise, provides HPE Edgeline Field Application Services, which help customers plan, design, build, and run IoT, edge and cyber-physical systems to accelerate deployment and ensure reliable and secure operation. These services include the evaluation of use cases, proof of value, solution deployment, and management of ongoing operations – helping customers get the most from OT/IT integrations.
Moreover, HPE Pointnext can help customers develop their own data acquisition, industrial network, and control components for HPE Edgeline OT Link Platform to create custom solutions and competitive advantage. HPE Edgeline OT Link Platform based solutions can be delivered on-premises with a turnkey deployment service, operated by HPE Pointnext.
Finally, HPE Edgeline EL300 Converged Edge System will be added to HPE GreenLake Flex Capacity, to deliver a consumption-based experience with usage-based payment, capacity metering, and tailored support, for customers who need a cloud-like experience for systems at the edge.
by Gary Mintchell | Nov 23, 2018 | Automation, Networking, Standards, Technology
I will only be at SPS for a few hours this year to check in with old friends and see some of the latest automation goodies. But I’m glad to be there at all. Thank you to Siemens who is sponsoring a press tour that includes a couple of days of intense cybersecurity briefings and workshops.
Oh, and a trip to Allianz Stadium to see the technology and a Bayern Munchen football match.
Some early SPS news:
- Avnu Alliance Demonstrates New Conformance Test Reference Tool
- OPC Foundation promises much news plus addition of Rockwell Automation
OPC Foundation
OPC Foundation has sent a couple of emails inviting us to a press briefing at SPS promising much news. I won’t be in Nuremberg on Tuesday, but I’ll catch up with Stefan and Tom for sure on Wednesday.
The mating dance has ended after a few months. Rockwell Automation has rejoined the OPC Foundation and gained a board seat. OPC Foundation has elected Juergen Weinhofer, vice president of common architecture and technology for Rockwell Automation, to its board of directors. Note that Weinhofer is also the Rockwell delegate to the ODVA board.
Weinhofer’s election to the board extends Rockwell Automation’s engagement in the technical work of the OPC Foundation and its technical advisory council.
“OPC UA has become the dominant open protocol for machine-to-software and machine-to-cloud solutions, and it is becoming critical for companies deploying a Connected Enterprise,” Weinhofer said. “I look forward to helping the OPC Foundation become a leader in machine-to-machine applications and helping OPC UA users unlock more value from their production systems.”
This quote is from the OPC news release. We should note that “Connected Enterprise” (capitalized) is the Rockwell Automation theme. I also note while parsing the comment that Rockwell is still firmly fixed in the factory floor area where Weinhofer specifically states “become a leader in machine-to-machine applications.”
“Rockwell Automation is a proven leader in industry standardization and open information technologies,” said Stefan Hoppe, president of the OPC Foundation. “I welcome not just Juergen’s business and political skills on the board but also the increased technical and commercial contribution that the wider Rockwell Automation team will also bring to the foundation.”
Avnu Alliance
Avnu Alliance, an industry consortium enabling open, standards-based deterministic networking, will exhibit at SPS IPC Drives in the University Stuttgart ISW booth. Avnu Alliance, alongside ISW and Industrial Internet Consortium (IIC), will showcase the role of conformance test plans, testbeds and test reference tools in ensuring an interoperable ecosystem of Time Sensitive Networking (TSN) devices.
“We are in cooperation with IIC, IEEE, IEC and others in creating an interoperable ecosystem through a common network foundation that stems from industry open standards and testing,” said Todd Walter, Avnu Alliance Industrial Segment Chair. “The market will continue to require multiple application layer protocols for networked industrial systems. The Avnu Alliance charter is to enable interoperability at the network layer, to ensure ‘One TSN.’ We are the organization focused on providing TSN test plans and reference test architectures to anyone in the industry that wants to test for TSN compatibility.”
As such, Avnu serves to support Fieldbus organizations by providing its TSN conformance tests and procedures to ensure those organizations’ interoperability in the wider Ethernet system.
Leveraging the industry-defined requirements for TSN network interoperability, Avnu ensures there is a universal set of test plans for conformance to guarantee interoperability at the network layer. Avnu has developed a baseline test plan in the industrial market that ensures industrial devices, whether end device, infrastructure component or silicon, conform to the relevant IEEE standards, as well as the industrial automation profile being defined by IEC/IEEE 60802 Joint Project working group.
Starting with Time Synchronization, or 802.1AS as the foundation for all TSN devices, Avnu released the first set of test plans at SPS IPC Drives in 2017. Avnu will soon publish additional conformance test plans for end devices, such as enhancements for scheduled traffic.
At SPS IPC Drives 2018, Avnu Alliance will show a new proof-of-concept (POC) Conformance Test Reference Design that offers a single, streamlined way for vendors to test TSN interoperability. The POC Conformance Test Reference Design is designed to automatically test TSN devices for compliance to 802.1AS. The demonstration features a Linux open-source test tool created by ISW in partnership with Avnu. This tool would also allow other protocol organizations to test application stacks on top of a TSN network in a streamlined way enabling one-stop certification at any test house.
by Gary Mintchell | Nov 9, 2018 | Automation, Industrial Computers, Internet of Things
Advantech has been appearing on a variety of lists of prominent Internet of Things suppliers. The Taiwanese computer company with a US office in Cincinnati, OH and intellectual leadership, supplies intelligent I/O, a variety of computing devices, and HMI devices.
Several years ago I was privileged to be invited to Suzhou, China to attend Advantech’s user conference. It was an impressive event. This year they called it the “first IoT Co-Creation Summit.”
More than five thousand Advantech clients and partners from around the world attended the summit. Here Advantech introduced its newest IoT platform structure WISE-PaaS 3.0 and 32 IoT solution ready packages (SRPs) co-created with software and industry partners.
The event in itself will aid in the software/hardware integration for various industries, connect and build a complete industrial IoT ecosystem and value chain, and allow Advantech and partners to officially step into the next IoT stage.
Advantech Chairman KC Liu stated that in view of IoT application characteristic’s diversity and fragmented market, Advantech has assisted industries in integrating and connecting existing hardware and software and regards creating a complete industry value chain as its primary task in IoT industry development.
Advantech is introducing new features for its WISE-PaaS 3.0 and sharing a number of IoT solution ready packages (SRPs), based on WISE-PaaS, developed with numerous co-creation partners. The company is also outlining future co-creation strategies and schedules for the upcoming year.
Allan Yang, Chief Technology Officer at Advantech said, “While IoT is currently flourishing and many companies have invested in connectivity and data collection equipment, we are still in the early stages of generating value from IoT data. Since WISE-PaaS launched in 2014, Advantech has continued its integration and improved connectivity with open source communities. Our IoT software modules are developed to create operational cloud platform services oriented around the commercial value generated by data acquisition. Data-driven innovation has thus become the main target for our WISE-PaaS evolution.
WISE-PaaS 3.0 offers four main function modules:
- WISE-PaaS/SaaS Composer: a cloud configuration tool with visible workflow. WISE-PaaS/SaaS Composer supports customized component plotting for simple and intuitive 3D modeling application and interaction. It updates views at millisecond rates and, together with WISE-PaaS/Dashboard, presents critical management data in a visually intuitive display to help extract valuable data and improve operational efficiency.
- WISE-PaaS/AFS (AI Framework Service): an artificial intelligence training model and deployment service framework. The WISE-PaaS/AFS provides a simple drag and drop interface that allows developers to quickly input industrial data. When combined with AI algorithms, the service builds an effective inference engine with automatic deployment to edge computing platforms. AFS offers model accuracy management, model retraining, and automated redeployment. It simultaneously controls multiple AI models in the application field; offering automated model accuracy improvements and life-cycle management services.
- WISE-PaaS/APM (Asset Performance Management): an equipment network connection remote maintenance service framework. WISE-PaaS/APM connects to a wide array of on-site industrial equipment controls and communication protocols. It supports the latest edge computing open standard, EdgeX Foundry, and includes built-in equipment management and workflow integration templates. Jointly with the AFS, APM accelerates Machine to Intelligence (M2I) application development.
- Microservice development framework: WISE-PaaS contains a micro service development framework to help developers rapidly create program design frameworks while reducing development requisites. Micro service functions, such as service finding, load balancing, service administration, and configuration center, all offer built-in flexible support mechanisms.
Advantech recently established a water treatment system, jointly developed with GSD (China) Co., Ltd., and a CNC equipment remote operation service, jointly developed with Yeong Chin Machinery Industries Co. Ltd. Both partnerships demonstrate how industrial digital transformations, led by Advantech and its partners through the co-creation model, offer innovative win-win IoT solutions.
Advantech’s IIoT iAutomation Group has launched a broad selection of rackmount GPU Servers from 1U to 4U. The SKY-6000 GPU server series are powered by Intel Xeon scalable processors and each of these highly scalable GPU-optimized servers support up to five NVIDIA Tesla P4 GPUs. IPMI management functions and smart fan control ensure better temperature control and thermal management environments. Every GPU pair includes one high-speed PCIe slot for highly parallel applications like artificial intelligence (AI), deep learning, self-driving cars, smart city applications, health care, high performance computing, virtual reality, and much more.
AI Deep Learning GPU Solution
With support for up to five pcs of half-length half-height (HHHL) GPU cards or one full-height full-length (FHFL) double deck card, plus one full-height half-length (FHHL) GPU card, the SKY-6100 series are designed for NVidia Tesla P4 HHHL GPU cards, making it the best choice for deep learning applications.
IPMI Server Management
With IPMI 2.0 support, the SKY-6000 series allows users to monitor, manage, and control servers remotely and receive alerts if any sensors detect device or component faults. In addition, event logs record important information about the server which can be controlled remotely using the IPMI KVM.
Smart Fan Control
The optimized thermal design separates the CPU and GPU fan zones, making sure the GPU card is not preheated or thermally affected by any other heat source. Also, with the smart fan control mechanism, fan speeds are controlled based on different CPU and GPU workloads and ambient temperature. This feature lowers the acoustic noise of GPUs that have heavy loading but not CPUs. Advantech’s SKY-6000 server series are available for order now.
