May 2, 2017 | Written by: Karen Lewis
Share this post:
Improve manufacturing operations and processes with adaptive robotics
Intelligent assets and equipment utilize connected sensors, analytics, and cognitive capabilities to sense, communicate and self-diagnose issues in order to optimize performance and reduce unnecessary downtime. Combining and analyzing disparate information from workflows, context, process, and environment, organizations can improve quality and optimize operations.
At the Hannover Messe IBM Booth, Benedikt Krueger, a member of the IBM Industry 4.0 Core Tech Team, walked me through the KUKA adaptive robot demonstration that enables technicians and operators to interact with it.
Photo: Adaptive robotics with KUKA
Integrate edge devices – such as KUKA robots – in the cloud
In an Industry 4.0 setting, directly integrating edge devices – such as machines and robots – in the cloud using Watson IoT Platform enables manufacturers to develop personalized products and services, improve operations, reduce costs and avoid the risk of downtime. By accessing different Watson services, in addition to other APIs on IBM Bluemix, a technician or operator is able to take advantage of analytic functions, predictive maintenance, and visualization of information in a dashboard.
In the demo, the reference architecture Benedikt uses is based on an Enterprise Service Bus (ESB) approach which enables him to integrate different machines with enterprise applications and manufacturing execution systems. An ESB integration strategy gives Benedikt the flexibility to determine which data is sent and processed in the cloud, versus which data is used locally. In this instance, the scenario being used enables the integration to occur directly in the cloud.
Enable different interactive services to assist machine technicians and operators
Benedikt explains: ‘Within the cloud we can implement different use cases, for example a chat bot supporting a service technician, or an operator responsible for the machine. The services provide the operator or technician with the ability to engage with the machine using natural dialogue.’
Figure 1: Predictive maintenance and cognitive repair use case
The specific dialogue and questions the KUKA operator or technician is interested in may include things such as: ‘Show me the information that pertains to a specific part of the robot;’ or, ‘Show me an overview of the parts of the machine;’ or, ‘Let me see the status of production;’ or, ‘What is the quality of the shift?’.
Using Watson, the responses retrieved are the best possible matching answers, found by searching a combination of structured and unstructured data comprised of repository information such as manuals, real-time operational data coming from the machine and its sensors, plus contextual information about the environment or operating conditions in which the machine is functioning.
The advantage of analyzing data in the cloud
Analyzing the data in the cloud makes it possible to access different services relevant to different use cases – all using the same dashboard. For example, visual inspection which enables the operator to use Watson’s image recognition service to monitor the quality of a product; or, voice and text recognition services which enable a technician to maintain machines by querying Watson IoT about a specific problem or issue related to the individual machine.
After a technician or operator speaks or types the problem into the dashboard, Watson offers condition-based recommendations to resolve the issue, providing relevant information such as repair instructions, machine manual pages, past maintenance reports, suggested lengths of time to carry out a repair, or success rates associated with the repair.
Once the repair process is started, the technician continues to receive additional information, and can also troubleshoot the problem if it is not resolved. The service technician can even leave feedback about the repair experience, enabling other colleagues to benefit from insight derived from the current experience – in addition to augmenting that machine’s existing base of information.
A common dashboard regardless of vendor or machine type
For service technicians responsible for repairing many different machines, and shift officers, who need to ensure all machines are running smoothly, it not unusual to encounter multiple machine interfaces that vary by the type and brand of equipment being used within the factory. Having a common dashboard – irrespective of the vendor, can streamline tasks and processes – from operations to maintenance right through to reporting. In Benedikt’s demo, the information is displayed and monitored through a common dashboard – across different machines, regardless of vendor.
Create a health-score from aggregated operational and sensory data
The dashboard monitors information such as the movement of the robot along different axis, the frequency of the engine’s movements, in addition to different sets of sensor data. Using input from Watson and the machine, the dashboard displays real-time operational and sensory data, aggregated as key performance indicators (KPIs) to create a health score for the machine. The combined information on the dashboard informs the technician about the machine’s health score which allows the employee to detect when a machine is likely to experience an operating error. The health score can also be used to schedule predictive maintenance, or to implement counter measures that mitigate potential machine issues in the near future.
Strike the optimum balance between operations and maintenance
Watson IoT Platform provides a spectrum of analytics that enable organizations to start quickly and realize value immediately and then apply more advanced analytics as more data is collected, while improving analytic skills. Watson analytics support a progression of capabilities to cover a wide variety of devices and assets from light bulbs and elevators, to robotics, trains and aircraft. The amount of time and energy you put into understanding the data and deriving insights will vary across the different types of assets.
IBM Maximo, IBM Predictive Maintenance and Quality, and IBM Watson IoT Platform can generate insights from the Internet of Things to help focus maintenance resources and obtain the greatest value from manufacturing equipment and assets.
Turn a KUKA industrial robot into a cognitive machine
KUKA industrial robots are used in a number of application areas, such as material handling, loading and unloading of machines, palletizing and depalletizing, spot and arc welding. They are used in a number of large companies, predominantly in the automotive industry, but also in other industries such as the aerospace industry. KUKA’s lightweight robots are used to perform complex assembly line tasks, collaborate with humans, and provide real-time data into the manufacturing process, creating cognitive factory environments using Internet of Things technology.
In an industrial setting, the use of cloud and Watson IoT solutions offers operators and technicians more control over the management and maintenance of critical machines in the following ways:
- Use a common dashboard across all your machines.
- Gain visibility into the health of all your machines.
- Monitor the performance of machines in real-time using a range of input and output values – operational, conditional, environmental, sensored, structured and unstructured.
- Respond to maintenance issues before they occur.
- Improve the efficiency of your factory’s performance.
Infuse intelligence into machines and processes
By tapping the power of cognitive computing, organizations around the world are moving from IoT vision and proof of concept to strategic deployments aimed at driving real transformation. Cognitive solutions infuse intelligence into machines and processes – from the factory floor to the finished product. The result is a new industrial era defined by factories, machines and parts capable of self-assessing, triggering actions and exchanging information with each other, and with the people who manufacture and maintain them.
Please use these resources to continue exploring how cognitive computing can transform manufacturing using factory IoT data to create a detailed view of production, asset health and quality.