Recently in Geocoding Category

By David Loshin

business value drivers, typical usage scenarios, and some ideas for melding process with application for a synergistic lift.

Location is a critical concept in many industries, yet the importance of a standardized approach to managing location is often unnoticed. For example, for some client scenarios, the business driver is reducing risk. Insurance companies like to see their customer base (and therefore, their exposure to certain types of hazards such as floods or earthquakes) spread across different geographic regions. Financial institutions may be subject to different laws (with different penalties for violations) at different geographic jurisdictional levels for the protection of private information.

Healthcare providers need to ensure that protected health information is not inadvertently exposed by being sent to the wrong address.

In other scenarios, the business drivers are financial, such as focusing on customer acquisition, retention or cost management. In some cases, marketing budgets are allocated to local print and media advertising to grow the customer base. In other cases, reducing transport costs by optimizing the supply chain looks at distances between delivery points.

Either way, the underlying desire is precision and correctness in geolocation.

But note that precision and correctness are very separate ideas, and that difference underlies some of the main problems related to address and location management. Next post: Accuracy vs. Precision.

By David Loshin


But knowing that with some degree of precision we can map locations to their nearest geocoded location, let's think about aspects of a more general challenge: ensuring resolution of a provided descriptive address to an actual known address.

Let me clarify this a little. When I talk about a "provided descriptive address" I am referring to what an individual has presented as an address. And while another individual might be able to infer enough meaning from a presented address to make a delivery, the address might have misspellings, errors, or other variations that might prevent it from being adequately mapped to a specific geocoded location.

Aside from the other benefits we have already considered, transforming the address into a standard form will simplify the geocoding process. That transformation process leverages a few straightforward ideas, namely:

1. There is a representative model for "standardized" addresses with its accompanying formats, syntax, acceptable value lists, and rules.


2. An application is able to scan a non-standardized (or what I called a "provided descriptive") address, differentiate between the parts that are good and the ones that do not meet the standard.


3. There is a way to map the non-standard parts into standard ones.

In fact, all three of these ideas are doable, and over the next set of postings let's look at each one of these in greater detail.

 

By David Loshin

geocode.

But what happens if you want to find the nearest address to a specific geocode? Or if you are only provided with a location description (such as "the intersection of Main and Elm Streets") and you want to get a precise geocode?

More practically, let's say you are a logistics manager for a delivery service and you are interested in mapping the most fuel-efficient routes for your drivers? In any of these cases, there is a need to link geocodes to addresses (and vice versa) in a way that is accurate and trustworthy.

And this is easily doable with standardized addresses. If you think back to my first post in this series, I talked about the need for delivery accuracy as a key driver for the definition of an addressing standard.

Here, we can use the addressing standard to provide accuracy in geocoding. Data services provide databases that relate street segments to boundary geocodes, allowing you to interpolate a geocode given an address. As long as we can get a standardized address, we can find the geocode.

At the same time, that same data product can be used to reverse map a geocode to a street segment, and with enough precision on the latitude and longitude values, probably even map to a specific address.

So the value of address standards goes beyond just delivery accuracy. As more applications rely on location to add value, the need for address standardization and geocoding will also continue to evolve and grow.

By David Loshin


From a delivery perspective, it could refer to a mail box, or might refer to the front door. From a taxation perspective, it might refer to a land plot. In other contexts it might refer to the street edge of a driveway or the center of a rooftop. Essentially, each of these instances may be different locations that share the use of an address.

In fact there are many locations that do not map directly to an address. For example, a telephone pole, a storm drain, the edges of a runway, and a point along the shoulder of a highway are all examples of locations that do not have addresses. So how do we find them?

The answer is to use a geocode: a set of geographic coordinates including a latitude and a longitude value. Geocodes specified with enough precision provide a very good way to pinpoint a location.

And with the growing use of handheld global positioning satellite (GPS) devices (or GPS-enabled devices such as most newer mobile phones), a geocode is practically as good as an address for the purposes of delivery. In my next post, I'll look at linking addresses and geocodes, and where that all fits within the concept of address standards.