The rapidly progressing Artificial Intelligence (AI) technology offers a lot of new opportunities to businesses in the industrial sector. Smart machines that are capable of performing repetitive tasks with accuracy and self-correcting errors would be the perfect solution for any factory involved in large scale production.
At the same time, the Internet of Things (IoT) technology can improve efficiency, scalability, and connectivity for industrial organizations while saving time and costs. Companies have started applying the sensing technology to improve workplace safety and operational efficiency, while reducing the cost of unnecessary maintenance.
The implementation of AI and IoT together will give companies a competitive advantage and help build the data-driven businesses of tomorrow.
Potential of AI & IoT in the Industrial Sector
The combined application of the two technologies is expected to extend data analytics beyond answering questions we have today, to solve unexplored questions, quickly and effortlessly. Combined AI and IoT predictive modeling used by data analysts would allow businesses to transition from descriptive analytics (i.e., there is a problem) to prescriptive analytics (i.e., here is how to solve the problem.)
We discussed the power of AI & IoT for industries with Steve Fearn, Chief Technologist at Hewlett-Packard Enterprise Group (HPE). Fearn explained that manufacturing concerns can improve their production processes by employing machine learning to different phases of the assembly process.
Fearn gave the example of a video analytics tool that retrieves data from the manufacturing execution system and compares it with the video image that it sees on the assembly belt to ensure that products match customer orders.
Often, manufacturers create the same product for all customers and this uniformity creates long production runs and saves costs. However, more and more buyers are demanding highly customized products that differ in terms of size, style, shape, design, and color.
The application of IoT allows manufacturers to precisely create goods that are highly customized, yet produced on a large scale, similar to homogenous production.
The video analytics system can verify the slight variations in products. Quality assurance through video can also check for any misaligned products, missing information, or faulty products. Ultimately, video analytics allows for a very high level of quality assurance without the need trained individuals for the job.
Managing the Digital Transformation
While digital transformation with IoT and AI can automate answers to old issues and solve new questions, one of the main challenges is to managing evolution from current production systems to ones based upon AI and IoT. Overcoming the challenge will require a singular effort by all groups developing, deploying, and using the system. This requires a systematic approach of identifying the current situation, highlighting the challenges, finding ways to resolve problems, and delivering results that give businesses a competitive advantage.
The current system of data analytics allows businesses to collect a huge volume of information about their customers, products, or business processes. However, there is no common framework for gathering and structuring data in a silo that can be shared across all the departments of the organization.
This leads to a number of problems for businesses.
• Production Inefficiencies
Manufacturing businesses often miss opportunities for reducing the cost of production due to incomplete data sharing and analysis.
• Design Flaws
When the data about customer preferences or needs isn’t shared across all levels of production, the final product may contain features that are not needed, or it could be missing features that are required by the client.
• Marketing Opportunities Missed
When customer demand and expectation data is inaccurate, businesses often miss opportunities in the market by producing less than their marketing team can sell.
• Customer Response
The lack of shared data results in delays when responding to defects reported by customers. A delay could also lower the quality of customer service.
Businesses looking to apply the emerging technologies of AI and IoT would also need to consider impacts on current employees. As with any new technology, Industrialized IoT would drastically improve production efficiency, but it could also result in job changes or even job loss.
However, Doug Smith, the CEO for Texmark Chemicals, took a different and perhaps even more effective approach: envisioning technology as a catalyst to empower employees to specialize and do more, rather than merely replacing them. Doug Smith notes that when his company began to implement IoT for production of chemicals, he made sure that operators and other employees were involved in the transformation initiative from day 1. One of the most important parts of the IoT journey for Texmark Chemicals was the involvement of the people who actually do the work in the plant. Without their support, input and vetting, it would have been impossible to introduce the sensor-enabled assets and other IoT technologies into their business.
Texmark is seen as a leader in innovation and digital transformation – it is called The Refinery of the Future – for the petrochemical industry. The company is not just talking about implementing the AI and IoT, but is in the process of implementing cutting-edge technologies that increase revenue, efficiency, safety, security and productivity, all positive impacts to the organization and the bottom line.
One way to improve the data processing for industries is through edge analytics. Edge analytics improve the process of data collection and speed of analysis by performing these functions where the data is gathered.
Eddy Biesemans, the global account manager for Schneider Electric, explained that big data and edge analytics would allow more complex calculations to take place in real-time while removing the problem of latency, saving costs on bandwidth usage and increasing security.
For instance, consider an airplane manufacturer looking to enhance their product’s performance by flying it for 2 million miles. It would simply be lengthy and inefficient for the manufacturer to achieve this goal.
However, the manufacturer could install sensors on a hundred airplanes that collect and process different performance variables as and when the aircraft is flying. Instead of transferring data to the centralized analytics system, the analytics model would be executed as the data is generating during a test.
The first rule of statistics is plot the data, as much can be inferred by simply look at the data. Similarly manufacturing businesses can achieve great results by developing a framework that leverages the high volume of data stored in data silos across different departments. The central idea is to take the data generated by different departments, such as sales, production, purchasing, etc., and make it accessible to all the participants in the organization.
For an auto manufacturer, data would follow the design or product lifecycle. The data cycle would begin with R&D, prototype, and component designing. Batch and performance data would be captured during manufacturing and quality assurance, shared across organization. The output from these functions would be used by sales to alert customers of shipments as well as after sales service to improve support, especially issues related to specific batches. Data generated by customers could be looped back into R&D.
This would create a continuous, virtual loop where all the departments would be involved. The data sharing across different organizational departments is aptly named closed-loop manufacturing.
Closed Loop Manufacturing
Closed loop manufacturing has the potential to become the industry standard for manufacturing businesses. The goal of the framework is to capture complete data about products such as how they are used and in what context, performance, demand fluctuations based on pricing, as well as customer rating.
Closed loop manufacturing would give industrial businesses four main advantages.
• It would help businesses identify the features that are critical to producing components of a high quality.
• It means starting small but thinking big. A company can transform step by step without putting production at risk.
• Engineers would be able to carry out a powerful root cause analysis with the framework, as it encompasses data across the lifecycle, not just from design or production.
• The framework can be continuously improved as the results from one cycle are used as input for the next stage.
Manufacturing businesses can take three lessons from the developments of new business processes.
• Progress in AI and IoT will significantly change the production capabilities for.
• Data visibility across the organization will improve a manufacturer’s response to defects and quality issues, and promote faster response to customers.
• The closed-loop manufacturing will significantly improve production and transform businesses through continuous improvement.
The key take away is to think big, but start small and learn through the process - and surround yourself with the right ecosystem of partners.
31 oktober 2022 (online seminar op 1 middag)Praktische tutorial met Alec Sharp Alec Sharp illustreert de vele manieren waarop conceptmodellen (conceptuele datamodellen) procesverandering en business analyse ondersteunen. Waardevolle online tutorial ...
3 november 2022 (online cursus van 1 ochtend) Workshop met BPM-specialist Christian Gijsels over business analyse, modelleren en simuleren met de nieuwste release van Sparx Systems' Enterprise Architect, versie 16.Intensieve cursus waarin de belangri...
21 t/m 23 november 2022 Praktische workshop Data Management Fundamentals door Chris Bradley - CDMP-examinatie optioneel De DAMA DMBoK2 beschrijft 11 disciplines van Data Management, waarbij Data Governance centraal staat. De Certified Data Man...