The value of artificial intelligence in IoT analytics

In many parts of Asia, seasonal heavy rains bring flooding, destroying citizens’ property and livelihoods. In the past, city administrations, citizens and businesses could do little but protect against flooding and the potential illnesses it brought. And technologies such as the Internet of Things (IoT), machine learning (ML) and artificial intelligence (AI) may provide breathing room for more forward-thinking leaders.

This is the application of DKI Jakarta Provincial Government Flood Control System in Jakarta Smart City. The project was developed by Jakarta Smart City in partnership with the Jakarta Water Services Agency (DSDA) and aims to optimize flood risk management in Jakarta. The project involves using IoT, artificial intelligence and machine learning as part of an early warning system to combat flooding risks in cities.

As more organizations deploy the Internet of Things in commercial and industrial settings, the volume of data coming from these devices and sensors may have a significant impact on improving quality, operational efficiency, and in Jakarta It is of great significance to save lives and property from natural disasters.

The speed and accuracy with which IoT systems can respond to their environment is critical, according to Kenneth Koh, director of industry consulting at SAS Institute. However, because devices and other sensors in typical systems generate large amounts of data, traditional tools and methods can slow down the process of making sense of this data.

What is Artificial Intelligence Embedded Internet of Things?

Kenneth Koh: Processing data at or near the edge can make IoT systems more flexible and impactful. However, the quality of data-led actions is as meaningful as the quality of the data-based insights on which it is based.

The Internet of Things itself is not new to manufacturers. Manufacturers have been collecting and storing sensor data from machines for decades. Their value proposition lies in AIoT – analyzing data in real time at the edge, leveraging artificial intelligence and machine learning to increase efficiency and value.

By equipping IoT systems with artificial intelligence capabilities, a variety of structured and unstructured data can be processed at the edge. Deliver high-quality insights faster for systems to act on.

How Artificial Intelligence Embedded IoT Unlocks Business Value

Kenneth Koh: Artificial Intelligence Embedded IoT improves operational efficiency and productivity, At the same time, the cost is reduced. It also drives innovation to provide better customer service, better products and faster time to market.

Embedding AI in IoT devices enables edge computing, allowing the deployment of IoT systems where consistent 5G networks are unavailable. For example, logistics providers can use IoT sensors in their transport fleets to monitor the internal and external conditions of vehicles, even in remote areas of the latter routes.

In addition to edge computing, AI-embedded IoT leverages machine learning to develop actionable insights from the terabytes of data generated by IoT systems every day. In the example above, data collected from these sensors is sent to the cloud in real time, allowing technicians to troubleshoot vehicle problems more accurately and faster.

Manufacturers can also use these insights to predict when a specific factory system or piece of equipment will fail, allowing technicians to implement preventive maintenance. Proactively detecting faulty equipment saves valuable man-hours while reducing costly unplanned downtime.

In retail, insights from IoT systems can be used to determine the optimal price for a product and minimize disruption to its supply chain.

The role of machine learning in IoT analysis

Kenneth Koh: Machine learning is artificial intelligence embedded in IoT compared to other IoT Deployment advantages. The system can learn while processing the data generated by the sensors, using a variety of advanced analysis methods such as decision trees, random forests, gradient boosting, neural networks, support vector machines and factorization machines.

This saves businesses human time and experts in the organization. Without the need to train AI systems extensively, experts can focus on other critical tasks while non-data scientists can access, view and process the data.

Machine learning capabilities also increase the range of data that AI systems can access and process: visual images, text and even spoken speech, both online and offline. The increase in the quantity and quality of existing data increases the value and impact of the insights gained from it.

Combining these machine learning capabilities increases the speed and volume of data processing, enabling real-time actionable insights. This is crucial in many IoT systems.

How AIoT supports Jakarta Smart City: Leveraging SAS’s artificial intelligence platform, Jakarta Smart City is able to integrate multi-source data in real time and provide advanced analytics through IoT, machine learning and artificial intelligence technologies to provide emergency/disaster prediction capabilities and Optimize services to the public. The result is a flood emergency response that reduces flood risk in Jakarta.

Given that IoT is historically an operational technology, who should own IoT security?

Kenneth Koh: The introduction of IoT is ambiguous blur the line between enterprise IT and OT. Sensors and devices are connected to the network to create new systems and improve processes. At the same time, this convergence exposes traditional OT equipment and systems to threats they never faced before.

In fact, true device security is a combination of technology, process and best practices. Therefore, securing IoT systems should not be the exclusive domain of OT or IT teams, but should result in closer and more effective collaboration between the two.

However, this is easier said than done because IT security teams and OT security teams often don’t speak the same language, making it difficult to understand each other’s perspectives.

The distribution of responsibilities is completely different. Priorities often diverge, and regulations governing OT security and IT security are sometimes conflicting. Gaining an overview of all assets in a given environment makes it clear which assets and processes cannot fail under any circumstances.

By doing this, organizations can establish and practice unified cybersecurity to ensure data confidentiality, integrity, and availability.

Cite a best practice for IT and operations technicians to work together

Kenneth Koh: In manufacturing, data versus time Very sensitive. For example, if chemical concentrations in a process deviate from optimal concentrations, engineers may only have minutes to react to save tons of product.

In many semiconductor processes, engineers only have seconds to react. In this case, analytics needs to move to the “edge,” meaning data must be analyzed and decisions made on the machine or on the shop floor, rather than in the back office or engineering department.

This requires the ability to perform analytics wherever needed, such as on the machine, on the production floor, in the cloud or in the back office.

One of the main challenges is data silos. For organizations that have not implemented IT/OT convergence, there is a patchwork of unintegrated or partially integrated applications and enterprise systems. Without careful planning, introducing new data sources, such as IoT sensors, can compound the problem.

Implementing a data integration platform to connect IoT systems with an organization’s existing technology stack can break down silos between historical and future data while providing a single point of control Give all teams the same access. This ensures that IT and OT teams are on the same page, laying the foundation for better IT/OT integration. ​

The above is the detailed content of The value of artificial intelligence in IoT analytics. For more information, please follow other related articles on the PHP Chinese website!