Business Analytics Blog

Business Analytics Blog

Opinions expressed on this blog reflect the writer’s views and not the position of the Capgemini Group

Figuring Out the Best Way Home by Cathy Buchanan and Guy Baxter

Category : Travel
Last week, with a beautifully sunny bank holiday weekend ahead, I decided that I’d finally try out the 20 mile cycle ride from my flat in London out into Essex to visit my parents. It seemed like the perfect opportunity to find out whether the journey by bike is feasible or not. (… a weekend where a parental rescue mission wouldn’t be out of the question.)

So I turned to some maps to find the “best” route.

Having spent over a year with the full time job of plotting daily routes for a fleet of vehicles, I’m well aware that the “best” route is quite subjective and that the first thing required is to determine and agree the aim.  

On Saturday I had no one else that I had to agree my objective with, but I still had to decide what it was. Did I want to:
  1. Get there as quickly as possible?
  2. Cover as few miles as possible?
  3. Visit as many green spaces and / or lakes along the way as I could?
  4. Use  roads / cycle paths that I’d used before wherever possible, to avoid getting lost?
  5. Avoid all major highways?
  6. Include as many hills as possible to get the best work-out?
In any optimisation problem, you’ll find that determining a definition of “best” can be quite the challenge. When someone asks you to find the “best” way of doing something, it’s definitely worth working out with them what would constitute best, because their priorities may not be what you’d expect. I’m sure there are people reading my suggestions above who are thinking “what about ….?”, because they’d prioritise something I didn’t even think of.

In fact, you’ve probably noticed that people have a tendency to focus on finding solutions to problems before being really clear about what they’re aiming to achieve. I’ve often wondered if that’s because agreeing an aim can be at least as hard as selecting a way forward.

Anyway, to help me out, I turned to the Transport for London website, which I’ve previously found to be very good at plotting out suitable ways to journey by bike. They’ve recently changed their journey planner and now give three options for any cycle route – “Easy”, “Moderate” and “Hard”. There is no explanation of what constitutes each definition, which plays havoc with my detail-hungry brain, with the only obvious pattern being that the estimated route times decrease as the suggested routes get harder. For example, the only one of the routes which didn’t include cycling along the hard-shoulder (?) of a dual carriageway on my journey to Essex was the Moderate one, which made my choice for me, telling me that, at least in that case, objective # 5 (avoid all major highways) was the one I cared about the most.

Inspired by my successful journey on Saturday  (although route planners and maps can’t predict locked park gates), I thought I’d look at the impact that different aims can have on an individual’s solution choice.

Let’s consider the route between the Olympic Park in Stratford, East London and Trafalgar Square in central London. Now this is a route which, on the face of things, seems quite straightforward … you follow the bright blue Cycle Super Highway which starts at Stratford and finishes at Aldgate, then head straight on past Bank and along Fleet Street and The Strand.

TfL and google recommended route

It’s so much a straight line that the Transport for London website has it as both the “Easy” and the “Hard” option. The only difference is the estimated time … which suggests an assumption that if you want an easy ride then you’ll cycle at two thirds of the speed. Google maps agrees with the route, and fixes the route time somewhere in the middle.
However, this isn’t a very creative route and although it meets objectives # 1 (cycle harder to get there faster), #2 (it’s a straight line) and #4 (it’s the main route that all the buses follow and then major tourist streets in central London); it doesn’t meet objective #5 (this route is straight down the main bus route into London), which we saw was my key decision maker. So this may be optimal for a lot of people, but I’d much prefer not to be cycling in and out of buses.

Transport for London’s “Moderate” route, takes a twisty turny path through the green spaces to the North East of the map shown above, which meets my criteria nicely, particularly as it claims to take no longer than the Easy route which is a gentle ride straight down the centre.

To see how differences in the definition of best can change the recommendation, I’ve created a small example:

Cycling example

Imagine being the cyclist who wants to cycle at an easy pace. You have 6 options:

First Leg Second Leg Time (min) Distance (mi) Green Space Main Streets
Side Streets Tourist Streets 55 8.5 N Y
Side Streets River 60 9.5 Y N
Main Road Tourist Streets 60 8.0 N Y
Main Road River 65 9.0 Y Y
Park Tourist Streets 65 9.0 Y Y
Park River 70 10.0 Y N

Choosing the best route for you is totally dependent on what your main objective is. If you’ve chosen objective #1 (get there quickly), then the best option for you is to go along the side streets and then the tourist streets:

Time taken by route

If you’ve chosen objective #2 (travel the least distance possible), then the best option is to go along the main road and then the tourist streets:

Distance by route

If your preference is a combination of objective #3 (use green spaces) or #5 (avoid main streets) then the best options are either the side streets or park, followed by the river.

Green parks and main streets by route

So you see how important the objective is to determining the “best” route ….

Before the “figuring out begins”, here at Figure It Out, we always take time to figure out what it is that we are figuring it out.

About the author

Cathy Buchanan
Cathy Buchanan
Working in Operational Research for over ten years, as a consultant my goal is to help clients understand the value that data brings to their organisations. Before consultancy, I spent 5 years working in call centre design and am now very interested in promoting “the contact centre of the future”, utilising & combining multiple customer channels efficiently. My areas of specialisation are Optimisation, Simulation and Data Visualisation.

Leave a comment

Your email address will not be published. Required fields are marked *.