RATE YOUR CYCLE ROUTES
Sep 2014 - Dec 2014 / Semester 1, MA: Design Informatics
Groupwork (4 people)- Group Leader
Inverleith Neighbourhood Partnership is a local community plan in Edinburgh with several high level themes. Project “Rate Your Cycle Route” is targeted to one of themes in Inverleith Neighbourhood Partnership: encourage people walking and cycling.
“Rate Your Cycle Route” is a website that can rate for cycle routes in Inverleith based upon the data of six cyclability factors; the score of each road is calculated by an algorithm created and optimized by the whole group.
FINAL OUTPUT
Demo Version of Website “Rate Your Cycle Route”
WEBSITE DEMO
Project Desciption and Project Blog
PROJECT BLOG
A Video about the research background and using process for the website.
ANIMATION VIDEO
More information about data collection, research process and algorithm evolution.
PROJECT REPORT
RESEARCH PROCESS
Fast
Hackathon
We did a one-day fast investigation about the reasons why citizens in Inverleith don’t cycle by using some investigation methods, such as interview, field visit and literature review. We raised several possible proposals based upon the first investigation outcome.
Slow
Hackathon
More in-depth investigation for data (multi-authentications toward one kind of data)
Analyse, select and organize data (Analyse six different cyclability factors and create an algorithm to calculate score)
A website that can rate for cycle routes based on different factors
SLOW HACKATHON
There are lots of factors influencing how easy and pleasant it is to cycle in Inverleith.
These factors are subjective and objective.
How can we combine and quantify these factors in a way that is useful and informative?
1. Data Collection and Data Analysis
1. 1 VoteBox
There are some certain criteria/ factors when we try to evaluate a road whether it is suitable for cycling or not. We want to know which factors are most important to the cyclists, therefore we make a "vote box", and put it during the Inverleith Community Conference and the ground floor of the Evolution House for one week.
The using process of VoteBox is simple and can be finished in one minute. There is a question on the top of the box. We provided 7 choices for people, which are the factors which might influence the cycling. People choose 3 cards from them and throw them into the vote.
Question: what are the top 3 considerations when you are cycling?
Options: Scenery on the wayside, cyclelanes, signage & navigation for bikes, road smoothness, gradients & slopes of roads; traffic flow.
From the outcome of VoteBox, we find that some factors are more important than others, which are: Road Surface Smoothness; Traffic Flow; Road Marking.
1.2 Questionnaire
All of the questions are some basic questions toward cycling and can be finished by all the people (cyclists / non-cyclists).
This table is the results of one question in questionnaire. This question is to ask people to give a score of the cyclability factors we provide, ten is the highest score, which represents this factor is very important, one is the lowest score.
The result of cyclists is the same with the vote box, people care more about road marking, traffic flow, road surface smoothness. The situation is a little bit different from the perspective of Non-cyclists, but is also similar with the overall situation.
1.3 BigMap
We put a big map on the wall during the Inverleith community conference.
It’s a simple task full of fun and really easy to finish: use green dots to mark the cycle routes they like; use red dots to mark the cycle routes they dislike.
We also find an issue we overlooked before :The negative dots mainly focus on the intersections.
1.4 Participants field testing and evaluation
We got five participants to finish our “testing and evaluation” task in Inverleith.
Task for the participants:
- Ride bike on a road (selected by us).
- Press the counter button on the bike when they notice cycling signposts
- After cycling, fill an evaluation form rating the different elements of the roads.
2. Cyclability Factors
From the result of several data collection approaches, we choose six cyclability factors that people care about most.
We rate every cyclability factors in the range of 0-100. 0 means the worst condition, while 100 means the best condition. Then we adjust the algorithm by changing the weight of each factor.
2.1 Surface Smoothness
Judgement Criteria：
We use vibration data (vibrational frequency and amplitude) as the factor to measure the surface smoothness.
Data Source：
We use the vibration sensor on the bicycle to gather this data.
We acquire the acceleration data of three axes, X,Y,Z with the vibration sensor.
We use X,Y,Z to define Vibration Coefficient to reperesent the smoothness of the road.
Vibration Coefficient is the standard deviation of X,Y,Z. The larger the number is, the less smooth the road is.
Calcalation point (start from 100)
After calculating we found the range of the Vibrating Coeffient is 0-10. So we subtructed 5 points every 0.5 of change in the value.
2.2 Traffic Flow
As the speed and volume of motor vehicles on a road increases, the greater the safety problems faced by cyclists.
Data Source:
We use 8:00 am -9:00 am peak time traffic volume as the basic data for traffic volume and use the road speed limitation for the measure standard of the vehicle speed.
1.Use the traffic volume data from the Edinburgh City Council https://github.com/edinburghlivinglab/cyclehack/tree/master/CEPATS
2.Use the 20mph Network Consulation http://edinburghcouncilmaps.info/transport/20mphconsultation.htm
Calcalation point (start from 100)
Based on the data we collected, the average traffic volume on 8:00am-9:00am peak time are normally between 200-1900, so the calculation above is a reasonable approach for data discrimination.
2.3 Intersection
Judgement criteria：
We divide all the intersections into two types: Big intersections and small intersections.
Data source:
main roads and local roads can be found in 20mph Network Consultation http://edinburghcouncilmaps.info/transport/20mphconsultation.htm
Score=100-12.5*X-5*Y+6.25*Z
X: the number of big intersection
Y: the number of small intersection
Z: the number of items of big intersection
Every road has two main intersections. Each intersection accounts for 50 pints. Intersections of roads are negative for cycling. We substract points because of the intersections, but the items of big intersections can add points.
2.4 Road Marking
Definition:
The road markings are defined as lines, patterns, words or other devices, except signs, for controlling, warning, guiding and informing the cyclists.
Judgement Criteria: We use different elements like type, width and color as the basic to give the points. Each charactertistic represents different value of points.
Data Source: We will use the data mainly based on the field investigation, with the support of data from Edinburgh City Cycling Map and stree view of Google Maps.
2.5 Road Gradient
Definition:
How steep the road is.
Steep grades are a barrier and potential hazard for bike riders. Riders build up speed on steep descents and may have trouble stopping or staying in control. This may lead to serious crashes.
Judgement Criteria:
The average tangent value and max tangent value of slopes(percentage value)
Calculate Point (start from 0)
The Inclination of the steepest road in Inverleith is around 5%. We subtract 10 points every 0.5% in average. In some areas the slope may become steep rapidly. We subtract 1 point every 1% of inclination on max slope of road.
2.6 Cycling Signpost
Definition:
The number of signposts and how recognizable the cycle signposts are for cyclists. Signage is important while cycling, cyclists could be guided and reminded well by these facilities.
Judgement Criteria:
The number of all the signposts on streets and the number which participant do not notice.
Calculate Point (start from 0)
The max number of signpost we counted on roads is 5. “100 / 5 = 20” We add 20 points to the score of road for each signpost. if participate miss one signpost, 10 points will be subtracted. If the signpost is not easy to be noticed, it loses part of its availability. 10 points will be subtracted for each signpost.
3. Algorithm
Algorithm Calculate Example: Raeburn Place
2.Algorithm after VoteBox & Questionnaire: Proportion of each parameter is added.
1. Algorithm after Fast hack: Simply add each factor together.
3.Algorithm after participatant evaluation: Some adjustment to the each points.
WEB DESIGN
Click any cycle route in the Inverleith, the website would pop up a little window that contains the overall score and detailed score for each cycability factor. Users can also see the tips of this route written by others.
User can choose the cyclability factors they care and then let the system recommend the cycle routes which are suitable for them.
FUTURE PALN
Finish the information of all the cycle routes in Inverleith (or maybe larger area).
Cycle route recommendation would be based on the location of user.
Real-time traffic status
(connect with google, ect).