Let’s (not) get physical: How satellite AI can improve human work speedsavril 14, 2022
We are excited to bring Transform 2022 back in-person July 19 and virtually July 20 – 28. Join AI and data leaders for insightful talks and exciting networking opportunities. Register today!
As technology evolves to support a wide range of tasks, companies are increasingly relying on automation to help improve overall work efficiency and operations. Satellite analytics, specifically, is rapidly growing in popularity and helping businesses in a variety of industries including utilities, energy, mining, transportation, construction and more. In fact, SnapLogic released a report stating 81% of employees say AI improves their job performance. Satellites can travel around the earth at 17,000 mph, capturing hi-res images to provide companies access to historical data, increase safety and cost-efficient insights.
There have been many satellites launched since the first one in 1957, and many more are coming. The latest satellites provide high-resolution images that enable a more detailed assessment of ground assets and hence, smaller objects. The AI component helps upgrade images to even higher resolutions and is on a similar trajectory as high-definition TVs. Moreover, this allows older imagery with a poor resolution to be used alongside new data, a process known as time sequencing. Today, satellites are obtaining more multispectral, infrared, and Synthetic Aperture Radar (SAR) images which allow the AI to define specific vegetation species and assess data at night or through clouds.
Industries tend to rely on more traditional methods of completing responsibilities, but satellite AI-powered asset inspections and maintenance can eliminate manual labor at a fraction of the cost. They are built to execute difficult and time-consuming functions to help companies save not only time but potentially lives in a high-risk work environment.
Drones can also be leveraged to complete such tasks if they are equipped with high-tech cameras and software to overcome visibility limitations during the day or night. However, satellites still prove to be more reliable, as drones may not be able to withstand certain weather conditions. Speed is also a factor to keep in mind. While drones can travel between 45–75 mph, they don’t compare to the orbital speed of satellites.
Artificial intelligence delivers information faster and at a higher scale compared to humans. Think of Sputnik, the first artificial Earth satellite that orbited the planet in 96 minutes. Satellite AI can capture large areas of land quickly, in some cases once every hour, and provide detailed insights and predictions based on the images – something humans cannot do instantly. To put it into perspective, manual crews at utility companies can conduct approximately 10 miles of detailed inspections per day, and drones can inspect up to 5 miles per day. Satellites, however, can complete over 10,000 miles of inspections in one day — quick and easy, right?
With more efficient inspections through satellite images, data analysis (artificial intelligence) and planning (AI-powered optimization planning), companies save even more money and time. Having access to historical and real-time data on specific assets in a matter of minutes supports a company’s future and timely decisions. In contrast, the ability to retrieve data from a manual inspection for the same assets can take weeks or even months. The cost is also relatively cheaper compared to other sources. Between the inspection of worksites and analyzing key findings, the cost of manual labor salaries and hourly fees of contractors and field crews can add up. Satellite-based analytics can aid in the productivity of these workers, reducing the time spent on these tasks and allowing them to focus on other, higher-value parts of the job.
The work environment for manual inspections can be dangerous, but satellites help increase safety. According to Esurance, the chances of getting into a car accident are one in 366 for every 1,000 miles driven. Considering that statistic alone it’s not safe to send crews out to inspect miles of lines or other distributed assets. Many locations are also not easily accessible such as mountains, swamps, backyards, etc. The odds of an employee being injured in a satellite inspection are significantly decreased, with a likelihood of no injuries at all.
There are over a billion geographically distributed assets across core industries in the U.S. alone and the cost of operations and maintenance at electric utilities continue to grow. Without the right resources, companies continue to lose money every year from liabilities related to wildfires, unexpected power outages, and more.
Satellite AI doesn’t only benefit one specific industry but several that you might be surprised to find out about:
Utilities: Vegetation management is often the single largest line item, exceeding $100 million, in the annual operating budgets of many larger utilities. It has also been one of the biggest and most complex challenges for this industry. Satellites help automate their manual tasks, which allows the workforce to be more targeted and productive with limited time.
Energy: Automation is a rising trend in the energy sector and satellite technology benefits the day-to-day operations such as life-cycle planning, costs, maintenance, and performance of assets. For example, detecting methane gas leaks from oil well pads can track a variety of metrics like land sustainability.
Transportation: Due to the wide spatial distribution of roads and railroads infrastructure, many transport assets are exposed and vulnerable to natural hazards. This inevitably increases the costs for transport agencies and operations. The power of satellites assist operators to predict vegetation-related risks along distributed lines and help mitigate hazards in the event of natural disasters like floods and earthquakes.
Water and wastewater: Water treatment, distribution and supply networks are spread across thousands of miles and can make it difficult for workers to navigate the area. Satellite AI alleviates the issue by effectively monitoring and managing its critical supply chain.
Mining: The mining industry is more reactive to unpredictable situations than being prepared ahead of time. Providing predictive maintenance strategies and automating mobile fleet monitoring through satellite AI helps companies be more proactive.
Construction: The construction industry is fairly dangerous and solely used to be operated by manual labor but now technology enhances the safety of crew members by managing heavy equipment fleets more efficiently and optimizing overall performance.
How to put AI to work
In any industry, there are several ways to incorporate AI-based solutions effectively and efficiently. Companies should consider reevaluating tasks and implement AI into their day-to-day workflows. New roles and responsibilities can be created specifically for AI to complement team members, allowing them to oversee initiatives and put more of a focus on other things that don’t require technological support.
Moreover, 49% of employees feel that AI has improved their decision-making and accelerated time-to-insights. AI can be leveraged to analyze issues at a high scale and identify the best areas your team should address. Employees can then view the AI results and share real-time feedback with the technology for continuous improvement. We’ll continue to see artificial intelligence integrate into a variety of industries, but keep in mind that while it takes on the heavy lifting, it will only be made stronger by human insight and expertise. The combination of AI and humans is a powerful duo and will help take workflows to the next level.
Abhishek Singh is the cofounder and CEO of AiDash.
Welcome to the VentureBeat community!
DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own!
Read More From DataDecisionMakers