Contents
The rise of food delivery platforms has seen an increasing demand for dark kitchens — delivery-only kitchens with no brick-and-mortar presence.
When deciding where to locate your dark kitchen, travel time analysis can simplify the site selection process. For example, you can use travel time data to show detailed trade areas for your food delivery service, identify blackspots where current or potential customers aren’t being served and pinpoint where to open up next for maximum coverage of chain restaurants.
In this post, we’ll show you how to use travel time data to analyse the cost-benefit of relocating a dark kitchen space from one high street location to another, cheaper high street location. To do this, we’ll be using ArcGIS Pro and the TravelTime add-in, which allows for easy visualisation of site catchment areas based on travel times.
What you’ll learn:
- How to use travel time data to evaluate which location has the most potential for opening a dark kitchen
- How to create travel time catchment areas for delivery zones
- How to integrate population data to understand population reach
Analysing the cost-benefit of relocating a dark kitchen restaurant
We’ll be using an example use case of a restaurant considering whether to relocate its dark kitchen from Oxford Street in Central London to a less congested area in Streatham, South London.
We can calculate the cost-benefit of moving to a dark kitchen using the TravelTime ArcGIS plugin and readily available data on average real estate prices.
Whilst Oxford Street has a significantly high footfall, we realise that the dark kitchen space doesn’t need to attract people, it just needs to serve them. That’s why we have gone with a reasonably busy, albeit less central, high road in Streatham.
Step 1: Install the TravelTime ArcGIS plugin and add population data
To begin, open up ArcGIS and install the TravelTime ArcGIS plugin. Once installed, you will need an API key and password to use the plugin — you can get a free API key here.
Once you have downloaded the plugin, you will be able to access the basic features on the toolbar, as well as the more advanced features on the side toolbox, as shown below. For this tutorial, we’ll be using the basic features only.
Step 2: Create travel time catchment areas using different transport modes
Now we can use the TravelTime plugin to create our catchment or trade service areas. We would like to analyse how many people we can reach within a 20-minute drive time limit.
We’ll begin by creating our driving catchment areas using the Quick Time Map tool from the TravelTime platform banner:
Since the TravelTime plugin provides full multi-modal transport options (including walking, cycling, driving and public transport), we are able to select our transport mode.
You can create up to 3 catchment areas at a time, adjust the timezone, select whether you are leaving or departing from a point, as well as the time of day you leave/arrive from such a point. You can also decide to run the analysis by entering a street address, or by simply clicking on the map.
For this analysis, we will set the transport mode to ‘Driving’ as this is the more common delivery method. We will create one catchment area with a journey time of 20 minutes, and set the time as leaving at 6 pm - peak dinner time! The TravelTime driving model works on a peak and off-peak basis, so changing the departure time will affect the results you receive.
We run the tool twice, using the postcodes of the two sites we are considering:
Oxford high street trade area in green. Streatham Hill trade area in red.
Immediately, you can see that Streatham Hill offers a larger service area (in red) versus our Oxford Street space (in green).
Step 3: Add population data to your travel time analysis
For this project, we need to add some population data to understand the catchment areas and who we can reach within our 20-minute drive time limit. The 2011 UK population census data is freely available and easy to use for analysis.
By downloading the data from the UK data service website, we can easily add it to ArcGIS through the ‘add data’ function. Once you have added this data, it should look like this:
Now it's time to overlay the population data to discover the population reach of both service areas. While TravelTime may not offer this functionality itself, we can use another ArcGIS tool to do this.
First, we will go to ‘Analysis’ at the top of the screen. From here we will use the ‘Clip’ functionality. When inputting the parameters we will use ‘UK population data’ as our input feature, while the clip feature will be our 20-minute drive time maps we previously created. Remember to save the output feature with your desired file name.
Once we have done this, click ‘Run’ (you’ll need to run it separately for each location) and you should get the following results:
Now that we have our desired population areas to show our trade service areas for delivery, we can open up the data to see exactly what population is contained in each of the maps.
To do this, right-click on one of the clipped layers and press ‘Attribute table’. You should get something like this:
You can now export this as a CSV or Excel file to calculate the population using the ArcGIS ‘Table to Excel’ tool.
What the analysis shows
After calculating the total person population, we can see that our Oxford Street space had a population of 164,700 people that could be served by this 20-minute drive time. However, Streatham Hill could serve a whopping 416,300 people at the same time - nearly 3 times more than the Oxford Street location.
We can also compare real estate averages for this restaurant versus our proposed dark kitchen space in Streatham Hill to see which space may be cheaper.
Loopnet, a commercial property search website, shows us that the average price of retail units for rent in Oxford Street is £134.03 per square foot. This compares to £28.50 per square foot that you would find in Streatham Hill.
Using a basic sized restaurant/kitchen food plan such as Subway, where the average floor plan is 1,200 square feet, this would result in £126,600 in savings in rent alone per annum.
Conclusion
And there you have it: using the TravelTime plugin in ArcGIS Pro, in a few easy steps we’ve been able to perform a cost-benefit and site selection analysis to identify where to relocate a dark kitchen restaurant to reach the most amount of people.
This analysis was done using a combination of travel time and demographic data. From this, we discovered the difference and cost-benefit of relocating a dark kitchen to a lesser-known High Street location with a higher population coverage.
Finally, while we used ArcGIS in this tutorial, you can also run a similar travel time analysis in several ways, using Alteryx, QGIS or the TravelTime API.
If you’d like to use TravelTime for your location analysis, start a free trial today.