Data-Driven Process Monitoring, Control, and Optimization for Industrial Systems

Dear Colleagues,

Data-driven process activities are imperative in industrial operations. Monitoring processes uses temporal and real-time operational sensor data to circumvent complex mathematical physics models so as to track, diagnose, and optimize plant operations. Process monitoring balances machine learning and multivariate statistics to detect anomalies, identify root causes, and predict equipment failures. Industrial-level case studies may be drawn from chemical processing plants or smart manufacturing lines as they exploit massive streams of time-series data. Data-driven monitoring treats a production system as a “black box,” extracting patterns directly from the continuous flow of variables like temperature, pressure, and vibration. Data-driven process controlling for production systems involves using real-time data, machine learning, and advanced analytics to pace and automate production operations. It utilizes continuous feedback loops to maximize efficiency and predict system failures. Furthermore, data-driven process optimization also takes advantage of real-time sensor data, machine learning, and advanced analytics to continuously analyze, improve and fine-tune operations. Instead of relying on static, historical baselines or human intuition, it employs algorithms to maximize efficiency, predict failures, and reduce costs.

This Special Issue on “Data-Driven Process Monitoring, Control, and Optimization for Industrial Systems” will focus on novel advances in research by developing new approaches or explaining modern industrial systems through monitoring techniques (multivariate statistical process control, machine learning anomaly detection, fault detection and diagnosis) by relying on data acquisition, preprocessing, modeling/feature extraction and real-time monitoring. The purpose is to improve predictive maintenance, minimize downtime and maximize safety as well as to demonstrate scalability from power grids to pharmaceutical manufacturing. Controlling activities should implicate real-time monitoring and visibility by deploying IoT sensors. Predictive maintenance performance will be benefited by eliminating unplanned downtime. Advanced process control will dynamically adjust operational variables in real time in order to achieve peak efficiency against raw material variability and environmental disturbances. Quality control will contribute to defect reduction as real-time adjustments ensure that a production batch maintains acceptable tolerances, thus averting defective products and reducing waste. Optimization studies will highlight data integration from IoT sensors, SCADA, PLC and ME systems. Moreover, predictive analytics will be deployed to forecast system behavior using machine learning and prescriptive modeling to recommend exact control adjustments. Root cause analysis will isolate controls that incite production bottlenecks, quality defects and excessive energy use.

Topics include, but are not limited to:

• The development of models that use PCA and PLS methods in multivariate statistical process control in order to reduce high-dimensional sensor data to a vital few variables.
• Extensive applications of SVM and LSTM networks to detect faults and diagnose hidden deviations in dynamic and non-stationary process data streams.
• The development of models that allow instant tracking of various relevant KPIs such as OEE, energy consumption, throughput rates and cycle times.
• The development of techniques and methods that optimize utility costs and carbon footprint in dynamic energy management systems.
• The development of predictive maintenance methods to foster condition-based upkeep to minimize downtime and extend the lifespan of heavy machinery.
• The development and application of quality control techniques and methods that continuously track production parameters to detect anomalies in real time in order to improve product consistency and curb scrap rates.

Dr. George Besseris
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 250 words) can be sent to the Editorial Office for assessment.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Processes is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• data-driven monitoring
• control
• optimization
• machine learning
• production improvement
• industrial systems
• quality control
• preventive maintenance
• smart analytics
• data integration