Taiwan-based Weintek designs and manufactures human-machine interface (HMI) products. An HMI is a user interface or dashboard that connects an operator to a machine. The operator controls machine processes or monitors machine performance through a touchscreen interface, which means they no longer have to use manual switches or record machine performance on paper. Today, HMIs are often used to optimize the performance of machines in industries such as manufacturing, oil and gas, and energy.
Historically, HMIs were not connected to networks, so machine controllers couldn’t access the HMIs remotely. They had to be machine-side, physically inputting commands into the HMI. Furthermore, data on a machine’s performance stored by the HMI couldn’t be shared and analyzed easily. Controllers had to download the data to a portable storage device and transfer it to a networked PC or laptop. “Only when the data was uploaded to the network could a company begin leveraging a machine’s historical data and identify possible performance alarms,” says Mao Hsu, product manager at Weintek. “The delay impacted how quickly companies could respond to potential issues, impeding optimization.”
About seven years ago, Weintek looked to network its HMIs to give machine controllers remote access and instant visibility on historical data. The company wanted to make sure that customers could avoid the burden of managing any on-site hardware to support the newly networked HMIs. “Our goal was to develop a cloud-based solution, which we would call EasyAccess 2.0, that gave customers stable network connectivity without any management overhead,” comments Hsu.
Weintek looked at hosting the infrastructure for EasyAccess 2.0 in the cloud and began to consider cloud providers such as Microsoft Azure and Amazon Web Services (AWS). Hsu says, “AWS was well-known to us, and the cloud service provider has a good reputation in Taiwan. Also, a key factor for us at the time was the AWS Global Infrastructure with its Regions and Availability Zones. It meant we could offer a low-latency service to our customers around the world. Competing cloud providers didn’t have the same maturity of service.”
IT personnel at Weintek built a cloud-based infrastructure on AWS for EasyAccess 2.0 using Amazon Elastic Compute Cloud (Amazon EC2). Amazon EC2 instances ensured reliable performance for the infrastructure’s web and access management services. Amazon Simple Storage Service (Amazon S3) also provided cost-effective storage for historical data.
Two years ago, Weintek began to develop the AWS infrastructure with the support of CKmates, an AWS Partner Network (APN) Advanced Consulting Partner based in Taiwan, and created EasyAccess 2.0 and HMI-related services. The updated service would include Internet of Things (IoT) technology, reducing the cost of collecting data from the HMIs for customers. Hsu explains, “The MQTT connectivity protocol available in AWS IoT Core is lightweight, meaning we can collect data from thousands of machines using relatively little bandwidth.” Besides integrating AWS IoT Core into the AWS infrastructure, Weintek also wanted to integrate AWS Lambda to automate notification services and Amazon API Gateway to enable customers to connect third-party HMI management tools with EasyAccess 2.0. “CKmates helped us build a proof of concept for the AWS IoT Core service and ensure our adoption of serverless computing and the Amazon API Gateway followed best practices. The service from CKmates was professional.”
Weintek built the IoT capability into its HMI service faster and at lower cost by taking advantage of AWS IoT Core. Explains Hsu, “Because the MQTT protocol used by AWS IoT Core is generalized—unlike the MQTT protocols used by other cloud service providers—we saved more than a month of development time and expense.” AWS IoT Core ensures the connections to the HMIs are secure with all the data encrypted. It also gives Weintek the opportunity to tailor security policies to specific HMIs if requested by a customer.
Weintek is using AWS Lambda and the Amazon API Gateway, in combination with a third-party chatbot service, to assist its support team. Any issue with a customer’s EasyAccess 2.0 service triggers an AWS Lambda call which, via the Amazon API Gateway, raises an alert on the support team member’s computer screen via the chatbot. The team member then responds to the chatbot message. “It’s a really cost-effective and efficient way to raise support tickets and maintain maximum service uptime for our customers,” says Hsu.
In the past, it would take several steps and physical interactions to make a server call to the HMI for machine performance data to be sent to the cloud infrastructure. Now, through the AWS IoT Core integration, performance data goes from the HMI to the cloud in a continuous stream. “It was tedious for customers to collect performance data if they had hundreds of machines. With AWS IoT Core and MQTT, it is a fairly simple process. Customers can program their own IoT service to start reporting HMI directly to the cloud as soon as a machine is in operation.”
Customers can leverage the data to respond instantly to potential upcoming performance issues. Hsu provides the example of a coffee-roasting machine manufacturer in the United States that uses AWS IoT for predictive maintenance. The manufacturer’s personnel now have real-time visibility into each roasting machine’s performance via AWS IoT Core. “The company has reduced the chance of a machine breaking down and halting production because they can manage the equipment proactively,” says Hsu. “Without our service running on AWS IoT Core, customers wouldn’t have such a deep level of insight into their operations to optimize performance.”
To learn more, visit https://aws.amazon.com/iot/.