Digital transformation of the hottest process indu

Digital transformation of process industry

digital transformation in process industry is a sub part of the digital process and is also regarded as the fourth industrial revolution, namely industry 4.0. The focus of industry 4.0 is that more and more companies will carry out digital transformation in the future, and those companies that adhere to outdated business models will gradually be eliminated. In capital intensive industries, manufacturing assets are less likely to disappear, and it is more likely to be acquired by a company that has completed its digital transformation, and the name of the acquired company will disappear. In view of this reality, this article introduces what we need to know to start the journey of digital transformation

the four necessary components of digital transformation in the process industry are data collection, data integration, data analysis and workflow transformation in order of operation

data collection

the equipment should keep good discrimination. The cost of lubrication measurement has never been reduced. If there is no data, the consequences are unimaginable, and we need to make data-driven decisions. Generally speaking, we have much more data than the actual effective data. The process industry has process variable record data, instrument and valve diagnosis data, dynamic equipment health data, asset utilization or OEE data, environmental report data, laboratory analysis data, work order/maintenance data, raw material data, product quality data, safety data and market price/spot price data. These are usually stored in file cabinets, engineer offices, workstations, and multiple servers in different locations

data integration

process data is stored in the historical data system; Laboratory data is stored in LIMS; As mentioned above in CMMs, work order and maintenance data are collected and summarized in many places in an enterprise. What if all the data can be consolidated in one storage location? This is the concept of data lake

once the data enters the data lake, all authorized personnel can get the data. When collecting data, you don't need to know the relationship between the data in advance. End users define the relationship between data lakes when using data. In other words, data regulation occurs before exiting the data lake, not when entering the data lake, which makes the data lake very effective in processing a large amount of data. Another advantage is that the data lake is conducive to further exploration and mining of data. Even if we are not sure whether some data is useful at present, we only need to store it in the data lake. Once the system needs to call these data, its value will be reflected. Not everyone needs a data lake. Some users can browse and collect data from multiple locations at will. Users who do not have massive data to process and do not care about the speed and efficiency of accessing massive data will not see the value of the data lake. Of course, if you can solve the problem of data analysis in Excel, you don't need a data lake. However, when excel cannot handle a huge amount of data, your enterprise really needs a larger database and better analysis tools

data analysis

all the tools, technologies, methods and models we use to transform raw data into useful information constitute an analysis toolbox. Repeatedly polish the data until all secrets are revealed. Analysis is a complex and fascinating process. Its goals include finding problems, informing conclusions and supporting decisions. If you have ever used bar charts or trend lines, these are simple and effective visual analysis techniques. You structure data into groups of data and apply expertise to raw data to make it easier to use and to identify trends or differences. Because the potential relationship is too busy or complex for the bar chart, we began to use other analysis tools

when the relationship between data is known, we build it in a model, and we can use this model to predict. For example, we know that the ideal gas law pv=nrt is related to the properties of the ideal gas and some actual gas models. If we know the pressure, volume, substance and quantity (NR), but want to calculate the temperature, we can apply the ideal gas law model to get some good temperature predictions. Similarly, if I have vibration and temperature data of moving equipment, I can apply the analytical model to determine the health status of the equipment, because I understand the relationship between vibration frequency, temperature deviation and potential failure mode or wear mode

if you spend enough time building effective device models, you can combine them and run them in relatively real time, which means that the digital twin technology has been gradually expanded. Digital twins can include process dynamics models or reliability models, energy and material economic models, or any combination of all these models

workflow transformation

the biggest benefit brought by digital transformation to the process industry may be the change of workflow through data analysis. It is the analytical ability that allows automated processes to repeat tasks that were originally performed manually, enabling staff to focus on improving knowledge and skill space, rather than simply repeating known solutions. How is it presented in the workflow of process industry? Operation or production engineers can include periodic workflow in their responsibilities, as shown in the figure below:

another monthly production monitoring and reporting responsibility is shown in the figure below:

technologies and products that automate workflow have existed for a long time. However, except for large-scale and high profit industries, the cost of these technologies is prohibitive for all industries. Therefore, these workflows are still manual, even though they are well suited for automated processes. With Emerson PlantWeb optics analysis and other popular analysis tools, the original manual operations can be easily and cost effectively deployed as automated workflows. In addition, reliability engineering workflow can be automated, expanded or simplified in the same environment. The reliability workflow is shown in the following figure:

the value of classifying these models into automated operations is that the device health information can be liberated from static reports and become real-time state aware information, which can be used by all people who need it in relatively real time. If you need to know more about PlantWeb optics asset management platform, please pay attention to our official account @ AI is the ideal testing equipment Emerson automation solution for paper tube manufacturers, quality inspection institutions and other departments, and we will contact you as soon as possible

now is a good time to discuss the digital transformation strategy. According to ephen coveyan, the author of the seven habits of highly effective people, St fixture is also very demanding. When enterprises without a transformation strategy or a digital transformation roadmap evaluate the above examples, they can hardly see measurable value. We may hear such words: anyway, I have to pay those engineers full-time wages. It is worthless to reduce their workload, or I cannot gradually reduce the number of employees to support this new expenditure. This kind of thinking only considers the direct cost that is easy to measure, but ignores the opportunity cost. When your colleagues are busy with repetitive work, what engineering work has not been completed? Or to put it another way, if you have an engineer in your staff, what else can you accomplish

enterprises with a strategic transformation vision can understand the high value created when acquiring and embedding engineering knowledge. They understand that the speed of automated analysis means more profit, because less time is lost in the decision-making and communication cycle. Strategic leaders recognize that improving situational awareness can make better decisions faster, which is a competitive advantage. They understand that it is a cost and capability advantage to complete more work with the same number of employees. Decision making occurs in a cycle of observation and decision behavior (OODA cycle). If any entity (organization or individual) can handle this cycle faster than its competitors, it can gain cumulative advantages at each stage, which makes its competitors more and more backward

the benefits gained through digital transformation efforts should not be considered merely incremental. Generally, by reducing waste, rework, or downtime, the direct return of the transformation workflow is measured in higher productivity

back to the topic, data brings new sources of energy, so we can say that analysis helps refine and process petroleum into various hydrocarbon derivatives, such as fuels, food, plastics, paints, solvents, soaps and drugs. Analysis is the intelligent application of raw data, which is transformed into operable information. Data analysis is more profitable than oil processing because it has lower risks and lower capital costs