Big Data in Chemical Industry

Chemical engineering has grown significantly as a result of technical advancements and data use. Big data’s insight and integration have sparked a wave of innovation. Chemical engineers have been able to innovate new processes, boost efficiency, grow volume, and collaborate closely with other industries by interpreting big data.

Rather than thinking of big data as a thing, it’s more realistic to consider it as a process. Although much of this notion is based on a large amount of data being collected, it is the analysis and interpretation of this data that makes big data beneficial, especially for chemical engineers.

Because of the connectedness that big data provides, communication between chemical engineers and other professionals in related sectors such as computer science and medicine has enhanced. For greater efficiency, information is now shared in real-time.

Commodity chemicals, petrochemicals, refinery products, specialty chemicals, life sciences, and consumer products are all part of the chemical process sector. Manufacturing facilities range in size from small-scale, specialized products like life sciences or consumer goods to large-scale chemical and petrochemical plants. Many industrial units are geographically concentrated in one location in today’s petrochemical and chemical complexes. As a result of the economies of scale, incremental advances in energy efficiency, dependability, and safety would be enhanced. According to a recent poll conducted by Price water house Coopers, 88 percent of chemical industry executives believe that data analytics will be critical in maintaining a competitive advantage in the next five years. As a result, big data analytics is predicted to be a hot topic in the industry in the next years. The chemical process industry has a long history of being one of the first to adopt computer-based control.

Data has proved benefits in efficiency, reliability, and safety in the chemical process sector. The energy industry’s driving force is to meet energy consumption demands in a clean, low-cost, and long-term manner. To better forecast customer demands, enhance energy management, and reduce environmental impact, data-driven approaches are applied. On the energy supply side, big data analytics may be used to optimize electricity output, anticipate plant outages, and forecast energy demand. Understanding consumer habits gives useful information for modifying their behavior and, as a result, minimizing consumption on the energy demand side. This section focuses on how data-driven approaches assist conventional and renewable power, smart grids, and building energy management.

Forecasting, real-time problem detection, load classification, and identification of energy consumption trends are all examples of how big data analytics is used in smart grid management. Smart grids make it easier for power generation, transmission, distribution, and demand management to share information. Smart meters collect higher-resolution data in real time, such as device status and electricity use (i.e., 15- or 30-min intervals). Smart meters can monitor and manage home appliances, as well as interact with other meters, giving users greater control over their energy consumption. Heat maps, 3D load graphs, and geographic information systems are effective for spotting concerns in energy consumption, and visualization plays a significant part in identifying trends and interpreting information from vast amounts of data.

Big Data in Chemical Engineering

The new era of Big Data (BD) is influencing the chemical industries tremendously, providing several opportunities to reshape the way they operate and for shifting towards smart manufacturing. Given the availability of free software, and the large amount of real-time data generated and stored in process plants.

why are many chemical industries still not fully adopting BD?

The industry is just starting to realize the importance of a large amount of data that they own to make the right decisions and to support their strategies. In this seminar we are exploring the importance of professional competencies and data science that influence BD in chemical industries for shifting towards smart manufacturing in a fast and reliable manner.

Chemical industries are manufacturing facilities that process various raw materials to usable end products using various unit operations such as distillation, extraction, separation, adsorption, absorption, chemical reaction, etc.

This includes processing facilities such as refineries, petrochemicals, pharmaceuticals, ferrous and nonferrous industries, specialty chemicals, food industry, etc. It is expected that the industry is at the very early stage of adoption of BD and the full benefits are not yet realized. Often companies do not very much understand the concept of digital transformation and several myths are hindering its path to success and value creation. It is noted that digital transformation using the latest technologies such as BD provides all kinds of industries several opportunities for value creation.

These technologies are also reshaping customer expectations and helping companies to achieve those expectations .

A few questions raised to the industry at this stage

1. Include the availability of digitally literate supervisory and executive board?

2. Digital central to corporate strategy?

3. Availability of platform business model strategy?

4. Leveraging existing capabilities and making big bets in new and advanced digital industry models?

Companies now are ready to drastically change their business models with new digital technologies like BD. This mainly involves modifications of the fundamental business operations and modifications in products and processes, organizational structures, etc . This article foresees to add value for researchers in the chemical industry to consider adoption of BD as one of the potential areas and thereby benefiting the chemical industry.

Recent improvements in data computing, storing and cloud technologies have enhanced computation capability, which in turn opens new opportunities for real-time monitoring applications such as smart unit operations, intelligent processing, autonomous operations.

Given the volume of process data such as flow, temperature, level, pressure, etc. generated by sensors in chemical industries every second along with other data generated from finance, maintenance, communications, etc. during normal operations, there exists a huge potential in chemical industries for benefitting from the adoption of BD.

It is important to understand that with the limited advancement so far, the role of BD in chemical industries goes far past the expected gains in efficiency, safety, and environment to achieve real-time analysis, better visualizations in real-time, generating alerts immediately, changing concepts about data storage and retrieval, etc. through the adoption of BD.

BD can effectively combine data from multiple operations in chemical industries such as process, maintenance, HR, etc, and provide real-time information about their assets and their operations which enhances the decisions making in real-time.

Such real-time decision-making in chemical industries helps to reduce manufacturing costs, inventory handling costs, and operational costs and thereby improving both productivity and profitability. Modern industries are emerging in the direction of large scale and complexity with many variables and with the widespread use of software and technologies, BD has been collected and stored. Therefore, maximizing the use of this BD for further improving the decision-making process is significant for complicated process monitoring systems used in chemical industries.

How can data science be applied to chemical engineering?

1. Improving process reliability by directly making operational recommendations. Think about thinks like improving your catalyst life, reducing improper operations (i.e. column flooding), improving consistent product quality. Machine learning algorithms have been applied in the process industries for decades to monitor faults, but with cloud technologies, the inclusion of unstructured data, and better data visualization, operational recommendations are being seriously improved.

2. Process data trending. Several startups are using data science to improve the way process engineers monitor process data. Essentially, what these companies are attempting to do is augment the classical process engineering workflow. For example, assume process event A happens. An operator classifies event A as a certain maintenance event. When the process engineer reviews data in the advanced trending client, the program will automatically classify modes of operation that appear similar

3. Business decisions. Business decisions in the chemical industry can translate to millions, sometimes billions, of dollars. By incorporating operational data, supplier data, and employee data, you can make decisions that optimize your business processes.

4. Remote monitoring. Wireless mesh sensor networks are being adopted for various applications (i.e. corrosion monitoring). Chemical engineers can help interpret the data and build applications that bring insights to engineers.

Pharmaceutical Industry

In the pharmaceutical sector, the primary topics in data use include speeding the pace of R&D and enhancing the manufacturing quality of end-products. Increasing R&D costs and stagnating product pipelines are putting pressure on the industry. The margins of existing pharmaceuticals and products have been steadily declining due to competition from generic drugs and expectations from healthcare providers and customers for lower-cost alternatives. Faced with these obstacles, big data analytics is being viewed as a potential source of significant additional value.

Food Industry

Sterilization methods (such as pasteurization, food packaging systems, preservatives, and irradiation) and reaction processes have all benefited from chemical engineering (brewing and fermentation). Knowledge of unit operations (distillation, mixing, fluid and solid transfer) has also enabled food production to be scaled up to an industrial scale. Furthermore, process automation and control have enabled efficient, large-scale, and high-quality food and consumable production. Advances in high-throughput experimentation, new sensors, data-driven modelling, numerical solvers, and optimization algorithms enable a new wave of computer-aided improvements in the food business in the big data era.

Energy Industry

The use of data in the chemical process industry has demonstrated improvements in efficiency, reliability, and safety. The driving force in the energy industry is to fulfil energy consumption demands in a clean, low-cost, and sustainable way. Data-driven approaches are used to better estimate consumer demands, to optimize energy management, and to reduce environmental impact. On the energy supply side, optimizing electricity generation, anticipating plant outages, and predicting energy consumption are examples of the application of big data analytics.

On the energy demand side, understanding consumer patterns reveals useful information to modify their behavior and, therefore, minimize consumption. This section focuses on how conventional and renewable generation, smart grids, and building energy management are benefiting from data driven approaches.

Challenges in chemical industry

The lack of a suitable software platform is a significant impediment to implementing and maintaining big data analytics solutions. The chemical engineering community collects data in scalar quantities (such as temperature, pressure, flow, and concentration), one-way arrays (such as spectrum, chromatogram, and particle-size distribution curves), two-way arrays (such as image and gas chromatography with mass spectrometry), three-way and higher-order arrays (such as video and hyperspectral images), and text data for a variety of challenges (such as email, operator log books, lab notebooks, and social media discussions). To make matters more difficult, all of this information is saved in a variety of places, including process historians, application and business databases, webpages, email memos, and handwritten notes. Using analytics to combine all of these data sources and draw meaningful conclusions is far from simple.

The future: more innovation in less time

Big data is comprised of three V’s: volume (amount of data), velocity (rate of data generation), and variety (data source growth). The three V’s of big data organically lead to the acceleration of innovation. More data and interpretation equals more growth, which then ultimately leads to even more useful data. That’s why an industry like chemical engineering is witnessing such expansive growth during our current data explosion: previous practices and methods restricted by old technologies have given way to real-time results, communication, and experimentation. Look for even more innovation in less time as the merger between big data and chemical engineering continues to mature.

References

1. ‘Adoption of Big Data by Global Chemical Industries’, by Ashiff Khan, A.Seetharaman, Abhijit Dasgupta
2. ‘Big Data Analytics in Chemical Engineering’, Leo Chiang, Bo Lu, and Ivan Castillo
3. What is Big Data — Characteristics, Types, Benefits & Examples by Abhinav Rai.
4. What is Data Science? (Oracle)
5. Big Data by Bridget Botelho
6. Adoption of Big Data by Global Chemical Industries (Authorea)
7. 5 V’s of Big Data (GeeksforGeeks)
8. What are types of Big Data by Richard Allen