You are here:  Home  >  Data Education  >  BI / Data Science News, Articles, & Education  >  BI / Data Science Blogs  >  Current Article

2018: Realizing Smart Industry Goals

By   /  February 15, 2018  /  No Comments

Click to learn more about authors Philipp Wallner and Seth Deland.

The industrial world is transforming with the rise of smart industry. Modern production machines and handling equipment have become highly integrated mechatronic systems with a considerable portion of embedded software.

The growing abundance of data is a key driver of smart industry. Power plants, production machines, electric and hydraulic drives, and vision sensors all gather an increasing amount of measured data during production operation. This data can be turned into business value by developing Predictive Models and algorithms. For instance, Machine Learning can train a model to understand historical sensor data, so that the model can be used to create systems to predict future equipment failures and prevent production downtimes.

But one major design question in implementing such systems is: where should the model be deployed? Should it run on the machine itself, so that decisions can be made in real-time on the attached programmable logic controller (PLC) or industrial PC? Or should it be in business IT systems, possibly located in off-site servers, or in the Cloud, where computational power is readily available and models can be easily updated? Each approach has its positives and negatives, and the answer depends on several elements.

Quite often, one of the first locations considered for Predictive Models are business IT systems. Model maintenance is straightforward, and should a better model be discovered, it is an easier task to update a model in this system than it is to update models in embedded systems. As a result, models can be continuously researched and improved as more data becomes available.

To integrate Predictive Models with such systems you need tools for integrating with a range of programming languages and APIs. Programming languages such as Java, .net, C++, and visual basic are often used to implement business systems. Other common data transfer methods include RESTful APIs and JSON. The generated Predictive Models must be able to integrate into these environments to avoid the costly and inefficient process of recoding models.

Software components are increasingly providing a significant part of the entire added value of machine or production plants. Embedded software running on PLCs, industrial PCs, or field-programmable gate arrays (FPGAs) involves closed-loop control functionality, ensuring product quality and predictive maintenance algorithms for increased uptime without service intervention. In addition, supervisory logic for ― in many cases even safety critical ― state machine handling and automatic generation of optimised movement trajectories are implemented in embedded software.

Response times are faster when implementing Predictive Models in embedded systems as data does not have to be transmitted over a network and back, and they are deterministic, running on a real-time system. In controls applications, where the result of the predictive model is used to calculate the next actions taken by the machine, this is especially important.

Implementing Predictive Models on equipment is part of a bigger trend to raise the complexity and size of the code base on production machinery. However, many machine builders are mechanically focused and must maintain experience with elaborate workflows and toolchains for mechanical construction. With regard to software design, machine builders rely on traditional methods for programming and testing on hardware, often being unaware of tools for modelling, simulation, automatic testing, and code generation which are widely used by their peers in automotive and aerospace industries.

Traditionally, algorithms tended to be developed by experts in IEC 61131-3, C/C++, VHDL®, or Verilog®. This was time-consuming and error-prone with an increasing complexity of the algorithms used in machinery. Manually implemented functions that have already been verified through simulations do not always behave the way they were intended to, may contain errors, and can cause missed deadlines and problems that are only noticed on-site.

Providing sophisticated sensor networks presents one of the vital prerequisites for achieving the efficiency, cost, and, therefore, competitive advantages promised by smart industry. To become innovative market leaders, equipment manufacturers must rapidly develop skills and expertise in new design approaches and technologies.

The design productivity and system reliability of such an approach can be improved by using Model-Based Design tools like MATLAB and Simulink. These tools facilitate modular development of automation components, hardware independent testing, and automatic code generation which can implement algorithms for specific hardware platforms with ease.

Real-time functionality is directly generated from simulation models using automatic code generation – thus avoiding sources of errors. The tested algorithm is directly translated into real-time code, saving time and allowing for the creation of innovative solutions in small development teams. Model-Based Design with automatic code generation lets engineers fully leverage their expertise in construction to build a machine or plant, minus the concerns about programming language details.

Central to the opportunities presented by smart industry and Industrial IoT is the collection and availability of machine data. In the future, engineers will be challenged to build confidence in using new methods and tools to overcome an ever-increasing amount of data and the growing complexity of software. In the meantime, industrial companies who can move their focus towards interdisciplinary design thinking (instead of production thinking) will rise out of the transformation with new business models for their market and as true innovative leaders.

About the author

Philipp Wallner in an industry manager for the industrial automation and machinery field at MathWorks, and is responsible for driving the business development of this industry segment that comprises energy production, automation components, and production machines. Prior to joining MathWorks, Philipp worked in the machine builder industry, where he held different engineering and management positions. He has a M.S. in electrical engineering from Graz University of Technology and an executive MBA in project and process management from Salzburg Management and Business School. Seth DeLand is application manager at MathWorks for data analytics. Before that, he was product manager for optimization products. Prior to MathWorks, Seth earned his BS and MS in mechanical engineering from Michigan Technological University. Follow Philipp, Seth, and MathWorks/MATLAB at: MATLAB Twitter, MATLAB Facebook, Philipp LinkedIn, Seth LinkedIn

You might also like...

Three Big Data Predictions for a Post-Hadoop World

Read More →