Issue link: https://resources.esri.ca/i/1313392
04 | GEOSPATIAL STRATEGY ESSENTIALS FOR MANAGERS MATTHEW LEWIN detecting potential hazards, spatial awareness is an important ability. For organizations, however, spatial awareness is important on a whole other level. The interests and assets of virtually any organization vary significantly by location. Whether it's customer spending habits or shipping routes, the factors that impact a company's fortunes differ in character and complexity from place to place. Spatial awareness is critical to the viability and long-term sustainability of essentially any organization ... or society, for that matter. Without technology, establishing spatial awareness is difficult, to say the least. The manual effort required to compile information about customers, competitors, suppliers, society and the environment, and then analyze and map the spatial relationships is arduous, complicated and time consuming. All this is to say, it's very costly. Geospatial technology overcomes this barrier. In simple economic terms, geospatial technology makes spatial awareness cheaper. Cost Drivers: Data and Machine Learning Before I talk about why cheap spatial awareness matters, it's important to understand what's driving the recent surge in demand. Geospatial technology has been around a long time, so this recent uptick suggests that certain new advances are accelerating the drop in the cost of spatial awareness. In fact, that's exactly what's happening, and two recent developments stand out: the explosion of accessible spatial data and the rise of machine learning. Humankind's current rate of data creation is staggering. Some projections have us doubling the world's data every two years, and by 2025 we could be doubling every 12 hours! That's a mind-boggling amount of information, and as mentioned, most of it is spatial. Several factors are contributing to the growth: • Advancements in traditional spatial data acquisition methods such as satellite imagery, aerial photography, surveying and remote sensing • Innovations in new acquisition technology such as UAV drones and vehicle trackers • Growth of in-situ sensor technology (the Internet of Things) • A massive influx of crowdsourced data generated by mobile mapping solutions and social media applications (including geotagged photos, videos and text messages) • The emergence of open data platforms from government agencies and businesses, allowing for the creation and sharing of raw spatial data, as well as derived works such as thematic maps and analytics dashboards, which themselves become sources of spatial data The net effect is that access to spatial data is byte-for-byte easier and cheaper than ever. And on that basis alone, we're seeing people and businesses take advantage. But data is only one side of the equation. The other side relates to how spatial data is analyzed. This is where AI and machine learning enter the picture. Machine learning gets around the problem of human involvement in analysis. I don't say that to sound flippant about human abilities, but certain aspects of spatial analysis are intractable— specifically, selecting variables. Machine learning uses prediction-based techniques to identify the most relevant variables to include in the model it builds. This is done based on how closely the variables produce an expected outcome derived from a training dataset. A great example is how insurance companies are using machine learning to identify property damaged in storms.