The document describes a project to predict occupancy trends in Barcelona's public bicycle sharing stations using open data. The project aims to help users plan trips by showing them which stations will likely have available bikes or empty slots. It uses a random forest model to make 3-day predictions based on past occupancy data, weather forecasts, calendar information such as holidays and weekdays. The model was trained on 90 days of recent observations and evaluated against data from the same months in the previous year.
Webinar: Using smart card and GPS data for policy and planning: the case of T...BRTCoE
2014/08/28 webinar by Marcela A. Munizaga
See more in:
http://www.brt.cl/webinar-using-smart-card-and-gps-data-for-policy-and-planning-the-case-of-transantiago/
Presentation by Lauren Sager-Weinstein, Head of Analytics, Customer Experience at Transport for London delivered as part of the joint BCS DMSG/DAMA joint event on 18/6/15.
Video version on YouTube at http://youtu.be/ZXMFumjWv2I
Transport for London’s data by its very nature is big— with 45 million bus journeys and 25 million Tube journeys each week being measured along with a variety of transport system data. TfL uses big data tools to combine these data sets to provide insight to our operations and for a better customer experience.
Sustrans Scotland Raising the Standards Day 2017: Monitoring and EvaluationSustrans
Our research and monitoring unit specialists explain how they can help you get the data to answer the questions of what you should invest in to achieve active mobility, by understanding the impact of infrastructure and behaviour change programmes.
Webinar: Using smart card and GPS data for policy and planning: the case of T...BRTCoE
2014/08/28 webinar by Marcela A. Munizaga
See more in:
http://www.brt.cl/webinar-using-smart-card-and-gps-data-for-policy-and-planning-the-case-of-transantiago/
Presentation by Lauren Sager-Weinstein, Head of Analytics, Customer Experience at Transport for London delivered as part of the joint BCS DMSG/DAMA joint event on 18/6/15.
Video version on YouTube at http://youtu.be/ZXMFumjWv2I
Transport for London’s data by its very nature is big— with 45 million bus journeys and 25 million Tube journeys each week being measured along with a variety of transport system data. TfL uses big data tools to combine these data sets to provide insight to our operations and for a better customer experience.
Sustrans Scotland Raising the Standards Day 2017: Monitoring and EvaluationSustrans
Our research and monitoring unit specialists explain how they can help you get the data to answer the questions of what you should invest in to achieve active mobility, by understanding the impact of infrastructure and behaviour change programmes.
Predictive Analysis of Bike Sharing System Using Machine Learning Algorithmssushantparte
Provided business solutions based on the ethical aspects of data collection and shortcomings of business by visualizing data and forecasting the demand using Ensemble Learning Technique (Random Forest) with an RMSE of 89.09%.
COVID19 Transport, travel & social adaptation study Wave 1 panel survey: inte...DecarboN8
COVID19 Transport, travel & social adaptation study Wave 1 panel survey: interim findings
University of Leeds, Institute for Transport Studies
Professor Greg Marsden
Professor Jillian Anable
Dr Llinos Brown
University of Stirling
Professor Iain Docherty
Mining dockless bikeshare and dockless scootershare trip data - Stefanie Brod...PyData
In September 2017, dockless bikeshare joined the transportation options in the District of Columbia. In March 2018, scooter share followed. During the pilot of these technologies, Python has helped District Department of Transportation answer some critical questions. This talk will discuss how Python was used to answer research questions and how it supported the evaluation of this demonstration.
We are Traffic: Creating Robust Bicycle and Pedestrian Count Programs, Krista Nordback, Ph.D.
As agencies looking to improve bicycle and pedestrian infrastructure have learned, it doesn’t count if it’s not counted. Counting provides information on the level of intersections, paths and roadways—data already available for motor vehicles but lacking for non-motorized travelers. For the first time, Federal Highway Administration’s Traffic Monitoring Guide now includes a chapter detailing how to monitor bicycle and pedestrian traffic. The slides from this webinar explain how to create a robust bicycle and pedestrian count program based on the new guidance. Agencies that show clear evidence of use are more likely to receive funding for projects, so join us and learn how to improve your existing count program or create a new one.
As smart data gradually become mainline data for transportation planning, some obvious flaws in infrastructure decision making become apparent when comparing traditional static data and the dynamic nature of human travel. The static survey, a common source of transportation, encouraged to assign a greater portion of longer trips and predicting more road widening and highways. In reality, shorter trips are dominant in cities. Shared mobility options could provide options for shorter trips. These short trips should be properly corrected and assign in our infrastructure projections when travel demand modeling is developed. Smart data is paving the way to open the door of a new possibility towards shared multimodal cities.
Master in Marketing and Communications - Marketing Plan and Digital Strategy ...Nikolas Dimopoulos
This is the presentation of the Final Project I devised for the Master in Marketing and Communications in Rome Business School, in 2019. It is a Marketing Plan for a Greek-Cypriot start-up company, called Mr Pengu.
Mr Pengu crafted a revolutionary platform, web and mobile-based, which allows customers to order almost anything, any time and from everywhere and have it delivered on their doorstep within 30 minutes.
The array of choices for delivery spans from groceries and pharmaceuticals to courier services and food.
Users are able to follow the delivery journey through the app or web-platform, thanks to the installed GPS devices across the delivery fleet.
Predictive Analysis of Bike Sharing System Using Machine Learning Algorithmssushantparte
Provided business solutions based on the ethical aspects of data collection and shortcomings of business by visualizing data and forecasting the demand using Ensemble Learning Technique (Random Forest) with an RMSE of 89.09%.
COVID19 Transport, travel & social adaptation study Wave 1 panel survey: inte...DecarboN8
COVID19 Transport, travel & social adaptation study Wave 1 panel survey: interim findings
University of Leeds, Institute for Transport Studies
Professor Greg Marsden
Professor Jillian Anable
Dr Llinos Brown
University of Stirling
Professor Iain Docherty
Mining dockless bikeshare and dockless scootershare trip data - Stefanie Brod...PyData
In September 2017, dockless bikeshare joined the transportation options in the District of Columbia. In March 2018, scooter share followed. During the pilot of these technologies, Python has helped District Department of Transportation answer some critical questions. This talk will discuss how Python was used to answer research questions and how it supported the evaluation of this demonstration.
We are Traffic: Creating Robust Bicycle and Pedestrian Count Programs, Krista Nordback, Ph.D.
As agencies looking to improve bicycle and pedestrian infrastructure have learned, it doesn’t count if it’s not counted. Counting provides information on the level of intersections, paths and roadways—data already available for motor vehicles but lacking for non-motorized travelers. For the first time, Federal Highway Administration’s Traffic Monitoring Guide now includes a chapter detailing how to monitor bicycle and pedestrian traffic. The slides from this webinar explain how to create a robust bicycle and pedestrian count program based on the new guidance. Agencies that show clear evidence of use are more likely to receive funding for projects, so join us and learn how to improve your existing count program or create a new one.
As smart data gradually become mainline data for transportation planning, some obvious flaws in infrastructure decision making become apparent when comparing traditional static data and the dynamic nature of human travel. The static survey, a common source of transportation, encouraged to assign a greater portion of longer trips and predicting more road widening and highways. In reality, shorter trips are dominant in cities. Shared mobility options could provide options for shorter trips. These short trips should be properly corrected and assign in our infrastructure projections when travel demand modeling is developed. Smart data is paving the way to open the door of a new possibility towards shared multimodal cities.
Master in Marketing and Communications - Marketing Plan and Digital Strategy ...Nikolas Dimopoulos
This is the presentation of the Final Project I devised for the Master in Marketing and Communications in Rome Business School, in 2019. It is a Marketing Plan for a Greek-Cypriot start-up company, called Mr Pengu.
Mr Pengu crafted a revolutionary platform, web and mobile-based, which allows customers to order almost anything, any time and from everywhere and have it delivered on their doorstep within 30 minutes.
The array of choices for delivery spans from groceries and pharmaceuticals to courier services and food.
Users are able to follow the delivery journey through the app or web-platform, thanks to the installed GPS devices across the delivery fleet.
Data Centers - Striving Within A Narrow Range - Research Report - MCG - May 2...pchutichetpong
M Capital Group (“MCG”) expects to see demand and the changing evolution of supply, facilitated through institutional investment rotation out of offices and into work from home (“WFH”), while the ever-expanding need for data storage as global internet usage expands, with experts predicting 5.3 billion users by 2023. These market factors will be underpinned by technological changes, such as progressing cloud services and edge sites, allowing the industry to see strong expected annual growth of 13% over the next 4 years.
Whilst competitive headwinds remain, represented through the recent second bankruptcy filing of Sungard, which blames “COVID-19 and other macroeconomic trends including delayed customer spending decisions, insourcing and reductions in IT spending, energy inflation and reduction in demand for certain services”, the industry has seen key adjustments, where MCG believes that engineering cost management and technological innovation will be paramount to success.
MCG reports that the more favorable market conditions expected over the next few years, helped by the winding down of pandemic restrictions and a hybrid working environment will be driving market momentum forward. The continuous injection of capital by alternative investment firms, as well as the growing infrastructural investment from cloud service providers and social media companies, whose revenues are expected to grow over 3.6x larger by value in 2026, will likely help propel center provision and innovation. These factors paint a promising picture for the industry players that offset rising input costs and adapt to new technologies.
According to M Capital Group: “Specifically, the long-term cost-saving opportunities available from the rise of remote managing will likely aid value growth for the industry. Through margin optimization and further availability of capital for reinvestment, strong players will maintain their competitive foothold, while weaker players exit the market to balance supply and demand.”
Levelwise PageRank with Loop-Based Dead End Handling Strategy : SHORT REPORT ...Subhajit Sahu
Abstract — Levelwise PageRank is an alternative method of PageRank computation which decomposes the input graph into a directed acyclic block-graph of strongly connected components, and processes them in topological order, one level at a time. This enables calculation for ranks in a distributed fashion without per-iteration communication, unlike the standard method where all vertices are processed in each iteration. It however comes with a precondition of the absence of dead ends in the input graph. Here, the native non-distributed performance of Levelwise PageRank was compared against Monolithic PageRank on a CPU as well as a GPU. To ensure a fair comparison, Monolithic PageRank was also performed on a graph where vertices were split by components. Results indicate that Levelwise PageRank is about as fast as Monolithic PageRank on the CPU, but quite a bit slower on the GPU. Slowdown on the GPU is likely caused by a large submission of small workloads, and expected to be non-issue when the computation is performed on massive graphs.
Predicting occupancy trends in Barcelona's bicycle service stations using open data
1. Gabriel Martins Dias*, Boris Bellalta, Simon Oechsner
Universitat Pompeu Fabra, Barcelona, Spain
Predicting occupancy trends in Barcelona's
bicycle service stations using open data
SAI Intelligent Systems Conference 2015
10-11 November 2015 | London UK
3. Bicing
The public bicycle system of Barcelona is called “Bicing” and it made
for local citizens. In order to use it, people have to pay an annual fee
that lets them borrow a bike for 30 minutes without any extra cost.
If the trip lasts longer than 30 minutes and less than 2 hours, a small
fee is applied. Otherwise, if it lasts longer than 2 hours the fee is much
more expensive.
Therefore, most of the trips (97%) last less than 30 minutes.
5. Barcelona
This is the map of Barcelona.
At the bottom, we can observe the sea. The terrain is not flat and the
highest altitude is over 100 meters above the sea level.
9. Problem
In general, the bicycles help people to travel around the city, to go to
work, to the school, and so on.
However, there are two problems which users face very often:
1. Not finding a bicycle when they want to go somewhere;
2. Not finding a free slot in the station when they arrived to their
destination.
19. Levels
Full
Almost full
Bikes and slots available
Almost empty
Empty
We observed that for a person that is looking for a bike, it does not
matter whether there are 5, 10 or 50 bicycles available in a station.
However, they want to avoid stations that may be nearly empty.
On the other hand, a person that is looking for a free slot will avoid
stations that are nearly or completely full.
Therefore, we defined such statuses as the critical ones.
48. 3 days of predictions
Using Random Forest
0
10
20
30
90 days of observations
49. 3 days of predictions
Using Random Forest
0
10
20
30
90 days of observations
50. 3 days of predictions
Using Random Forest
0
10
20
30
90 days of observations
51. 3 days of predictions
Using Random Forest
0
10
20
30
90 days of observations
52. 3 days of predictions
Using Random Forest
0
10
20
30
90 days of observations
0
10
20
30
30 days of observations - 1 year before
53. 3 days of predictions
Using Random Forest
0
10
20
30
90 days of observations
0
10
20
30
30 days of observations - 1 year before
54. 3 days of predictions
Using Random Forest
0
10
20
30
90 days of observations
0
10
20
30
30 days of observations - 1 year before
+
55. 3 days of predictions
Using Random Forest
0
10
20
30
90 days of observations
0
10
20
30
30 days of observations - 1 year before
0
10
20
30
Next 72 hours
+
56. 3 days of predictions
Using Random Forest
0
10
20
30
90 days of observations
0
10
20
30
30 days of observations - 1 year before
0
10
20
30
Next 72 hours
+
We considered the last 90 days and the 30 days observed 1 year
before to predict the statuses in the next 72 hours.
70. Accuracy
According to the age of the predictions
0 %
50 %
100 %
0 days old 1 day 2 days
Without open data Using open data
71. Accuracy
According to the age of the predictions
0 %
50 %
100 %
0 days old 1 day 2 days
Without open data Using open data
72. Accuracy
According to the age of the predictions
0 %
50 %
100 %
0 days old 1 day 2 days
Without open data Using open data
We observed that the average accuracy was improved when we used
open data in the predictions.
77. Sensitivity
Using open data
According to the age of the predictions
0 %
25 %
50 %
75 %
100 %
0 days old 1 day 2 days
& Specificity
The sensitivity is the percentage of critical statuses that were
correctly predicted. It was over 75%.
87. Future work
Scalability
Other data sources
Our future work involve making the predictions for all 400 stations,
considering other sources of open data.
91. Future work
Scalability
Other data sources
Smartphone application
Other applications
We expect that the city council might use the predictions to improve
their schedule to collect the bikes from the full stations.
92. Gabriel Martins Dias
gabriel.martins@upf.edu
Boris Bellalta, Simon Oechsner
Universitat Pompeu Fabra
Barcelona, Spain
Predicting occupancy trends in Barcelona's
bicycle service stations using open data
Impact of
external
factors
Critical
statuses can
be predicted
Use of
open data