Are Maps all Created Equal?

I really enjoyed reading Brian Rudman’s article in this morning’s NZ Herald. It was about Google Maps and the quality of their data in Auckland. Basically it was about the usefulness of Google Maps to help people find their way around, getting walking and driving directions, and most recently the inclusion of information to help people find out which buses to catch and how to get to them using data from ARTA.

I urge you to read the story, because it highlights some interesting points that we often struggle to explain to people.

I’m not knocking Google, I love Google and spend a lot of time using it, as do many of my colleagues. But here’s the thing. Google is a data collection and aggregation tool that enables people to access data from multiple sources and use it for their purposes. In some cases Google creates the data, which includes having people drive vehicles such as the Street View cars to help people make better use of maps.

The problem we frequently have is that people think that, because there are Google Maps and Google is ‘the authority’ then their maps must be the best, or, as people often learn the hard way, that Maps are all pretty much the same.

If that were the case, there would be no need for GeoSmart, because Google obviously has far more money and resources than we do. What we have and they don’t, is a mandate to have the best possible maps that can be used to meet people’s varying needs. One of the key components in this, is what we call our ‘turn restriction database’. We know where all the roads are, we know which ones have traffic lights or roundabouts, we know which ones are one way streets, or have no left or right turns. We know the streets where you can turn legally, but a large vehicle probably wouldn’t be able to complete the manoevre.

We know which roads in NZ actually exist. What do I mean? New Zealand was town planned in Edinborough a couple of centuries ago and some 20% of the streets draughted, were never constructed. We know those as paper roads. These still exist on our government maps (which services such as Google use) because they have a legal status and the Government can still retake the land to build them.

For decades, our people have maintained maps of New Zealand working with data we collected by driving and flying New Zealand over and over again. We continue to do this and move the boundaries taking advantage of new technology so that our data quality and accuracy stays ahead of the needs of our clients. For example, with the RAPIDcV, we now are able to collect data at 15cm accuracy for future car navigation and safety systems. Advanced Driver Assistance Systems, Eco-routing and other future technologies will only work with quality data and for these services, near enough is definitely not good enough.

When people buy car navigation systems, they are relying on accuracy to help them find their way around. There is a good reason why quality brands such as Navman and TomTom come to us for data, because near enough is not good enough. If you take a look at web map sites where you can see the roads on aerial photography and where they show the roads using the labels, you will see that they often don’t match up. In other words, they are not spatially accurate.

If you want to claim tax rebates for times when your commercial vehicle is not on a public road, you need to to be able to prove accurately, where you drove. If your map itself isn’t accurate, then your argument must be flawed.

In the old days, we looked at a map and interpreted the data in our heads. If something didn’t look right, we worked our way around it, and it wasn’t a problem. When you put your map on a computer and have the computer make decisions for you, the quality of the data has a far more serious impact. That is why we have a large team of professionals employed in NZ to make sure that we have as accurate data as possible. That is why the NZ Automobile Association invested in our company.

One of today’s problems is that these maps are now accessible on mobile phones and other devices. People assume that all maps are basically the same and then don’t understand when they get a poor result. They might blame the phone manufacturer or the technology, but the old addage in the computer industry is still true. GIGO. Garbage In, Garbage Out.

So next time you want to rely on a data source, don’t assume that all maps are the same. They aren’t. In some cases it doesn’t matter, but in many cases it does. Our people care about quality, they use patience and skill to produce map data that people rely on.

Thanks Brian for showing us that map quality matters and being a multinational giant doesn’t necessarily mean they are always the best. Who knows NZ better than Kiwis? I’m sure you will find AA Maps and other sites that use GeoSmart Maps a tad more reliable.

The Travelling Salesman Problem

GeoSmart now has a solution available for people wanting to know which order to visit their clients in, so that they can keep time and travel costs to a minimum. This is a form of Route Optimisation known as the Travelling Salesman Problem. GeoSmart also offers complex Route Optimisation under the banner of Route2GO, but that is another story.

With the price of petrol now permanently above NZ$2 a litre and time at a premium, this tool can easily assist people in making the most of their travel.  It could be a delivery truck working out the order of their deliveries and in reverse order how to load the truck. It could be a sales person working out the best way to make their sales calls.

In the illustration you can see that we set the first and last addresses as fixed. In other words we are defining where we start and finish for the day, they don’t have to be the same, the last one could be a motel yo are staying at for the night and the first one might be your home or office. You could just fix the start point and be flexible on the end point and then go to the AA Maps website to decide where to stay for the night and even make the booking before you leave.

In the illustration, the sales person guessed an order in which to do the calls, but the optimised route cut 34km from the route and at the (adjustable) per km rate resulted in savings of $10 on the trip.

This tool doesn’t just have to be used for a set of destinations. It could also be used as a trip/price calculator. For example one of our new clients is putting this tool (slightly modified so that you can’t edit the price per km which only he can do on his site) so that people can get their own quote on a delivery job without having to ring the company.

Many companies pay staff or contractors on the basis of an agreed distance, it could be for commuting, or for visiting clients, patients etc. This tool can calculate the ‘fastest’ route between clients and will deliver a consistent result that can reduce arguments and deliver an equitable result.

There are other standard features to this tool, for example click on any of the letters denoting locations and the map will instantly zoom to that location so you can see it better. The maps are dynamic and do of course have full pan and zoom functionality.

This optimisation is possible because GeoSmart has a full turn restriction database for New Zealand so when it calculates these directions it knows all about one-way streets, dual carriageways, no right turn, no left turn etc. This is in fact the same database used by te leading car navigation brands including TomTom, Navman, Siemens VDO, BMW, Ford, Honda,Horizon and Nav N Go.

If you have a route optimisation problem, please send us an email to info@geosmart .co.nz or follow the directions on our Contact Page.