The Industrial Internet of Things (IIoT) and smart manufacturing initiatives are accelerating the creation of new technologies, transforming global manufacturing. Like most innovation, competitive advantage and measurable, improved economics will be the biggest drivers of adoption. Enhanced sensors, data collection, analysis, visualization and interpretation will drive accelerated learning from existing processes and inform the design of new ones.
This closed loop will have the power to link the entire value chain from raw material to end user and will eventually lead to continuously optimized manufacturing driven by artificial intelligence. Specifically, in polymer manufacturing, a broad approach is underway to improve and optimize the production and research of new materials.
Synthetic and natural polymers make our modern lives possible through a multitude of industrial and consumer applications ranging from manufacturing and medicine to oil and gas, water management, food, electronics and construction. Continually pushing the boundaries of performance in existing and new applications is what drives the development of new materials. Additionally, there is a perpetual drive to improve the efficiency, yield and quality in the production of these materials. Typical examples of companies focused on implementing changes targeting improvement include information sharing, data analytics, enhanced process measurements, process modeling and control algorithms, and the training and re-training of factory operators, technicians, scientists and engineers.
For decades, much of the chemical manufacturing industry has operated through a continually-evolving combination of approaches including quantitative, empirical, fundamental first principles and qualitative processes. Empirical and quantitative analysis has been a major focus, evidenced by the chemical industry’s leadership in its use of process data.
The industry has a long history of recording and utilizing data to analyze performance, troubleshoot problems and incrementally improve existing processes. Fundamental chemistry and process understanding, good maintenance practices, and talented engineers have worked alongside these empirical methods to optimize processes, leading to active development and improvement of process modeling and control efforts. This leads to improved plant design incorporating state of the art technologies and methods.
Another common phenomenon is the qualitative experience of seasoned front line factory workers who have learned, managed and operated production assets for years. Replicating the intricate sensor network (sight, sound, smell, touch, taste) and the natural human capability of advanced pattern recognition and rapid historical recall is a very complex problem with the current sensor mix and capabilities found in many facilities. This is especially pronounced where processes are particularly complex and regularly evolve with the requirements of end users, such as in polymer production.
Some issues lie in complex chemistry where true quantitative or fundamental first principle understanding of production is rendered difficult or nearly impossible due to variability in plant-to-plant equipment and process design, conditions, feedstocks and other inputs. Typical problems then become inconsistent quality across a network of plants producing similar products, off-spec material, longer than required cycle times, or over-processing resulting in other quality issues.
Many industry standard methods rely on off-line measurements in addition to the combination of quantitative, empirical, qualitative and fundamental methods. However, as processes continue to evolve, so will the required technology and tools to operate them efficiently and effectively.
In the future, as manufacturing advances, so will the research, development and manufacturing of new materials. These materials, many polymer based, will drive next generation consumer and industrial products. Just as the manufacturing processes are engineered to be ‘smart,’ materials themselves are undergoing a similar transition to be ‘Smart,’ more customized and more application specific. This transformation requires a new fundamental understanding and tighter control of production assets.
The polymer industry, like many others, is already shifting to embrace and experiment with new technology and processes to drive the needed optimization and control. With IIoT, smart manufacturing and Industry 4.0, more sensors, data and real-time information about fundamental production will provide additional leverage to existing manufacturing systems. Examples can include the dynamic fusion of data sets from process sensors and new detector measurements to yield typically sophisticated data in a simplified format.
In the first instance of new systems, there will be human-in-the-loop control where real-time actionable information is provided to production personnel to optimize, correct disruptions and operate more effectively. Once critical parameters are measured, these can oftentimes be combined with existing and new kinetic models for predictive, artificial intelligence based closed loop feedback control.
Finally, this rapid evolution in many industries when simultaneously leveraged with robust wireless networks in manufacturing, advanced data analysis algorithms, visualization and simplified data handling will present a myriad of opportunities. Another major opportunity is the cross-pollination of ideas and technologies among seemingly opposite industries such as transportation, medicine and manufacturing, which could drive true transformative innovation and change. Often, people solve complex problems that are directly applicable to new problems in other industries, but a generally siloed approach prevents dissemination. This cross-pollination is an absolute necessity to take advantage of the rapidly changing landscape and will serve to accelerate the current evolution of manufacturing.