Using JMP® Visualization for A Bike-Sharing Program in NYC Discovery Summit 2014 Matt Flynn, Aetna; Mary Loveless, JMP Put titles in here.

MnM is an analytic consulting company. Team members are: Discovery Summit 2014 Who we are MnM is an analytic consulting company. Team members are: Matt Flynn: awesome SAS and JMP programmer; all-around genius Mary Loveless: storyteller and JMP user MnM looked at Citi Bike’s open, free data to understand patterns of ridership and bike usage in New York City. Tools we used: SAS R JMP

Citi Bike Data Data Collection Data Description Visualization Flow Discovery Summit 2014 Overview Citi Bike Data Data Collection Data Description Visualization Flow

This talk will focus on the New York City bike share program: Discovery Summit 2014 Citi Bike DATA Citi Bike offers a public bike-sharing service in many cities in the US. This talk will focus on the New York City bike share program: Membership Rider usage Availability: Bikes (access) Docks (returns) Station and neighborhood usage

The data (Citi bike public data) Discovery Summit 2014 The data (Citi bike public data) Bike Station Data is available in JSON format from https://www.citibikenyc.com/stations/json Contains mostly: Station Names and IDs Available Bikes and Docks Dates and Times Trip History Data available in CSV format from http://www.citibikenyc.com/system-data Contains information regarding bike trips, such as: Start Time and Date Stop Time and Date Start Station Name End Station Name Station ID Station Lat/Long Bike ID User Type (Customer = 24-hour pass or 7-day pass user; Subscriber = Annual Member) Gender (Zero=unknown; 1=male; 2=female) Year of Birth

The data is in a JSON file What is a JSON file ? Discovery Summit 2014 Where do we begin? The data is in a JSON file What is a JSON file ? Wikipedia definition: “JavaScript Object Notation is an open standard format that uses human-readable text to transmit data objects consisting of attribute-value pairs. It is used primarily to transmit data between a server and web application, as an alternative to XML.” JSON (JavaScript Object Notation) is a lightweight data-interchange format. It is easy for humans to read and write. It is easy for machines to parse and generate. It is based on a subset of the JavaScript Programming Language, Standard ECMA-262 3rd Edition - December 1999. JSON is a text format that is completely language independent but uses conventions that are familiar to programmers of the C-family of languages, including C, C++, C#, Java, JavaScript, Perl, Python, and many others. These properties make JSON an ideal data-interchange language. JSON is built on two structures: A collection of name/value pairs. In various languages, this is realized as an object, record, struct, dictionary, hash table, keyed list, or associative array. An ordered list of values. In most languages, this is realized as an array, vector, list, or sequence. These are universal data structures. Virtually all modern programming languages support them in one form or another. It makes sense that a data format that is interchangeable with programming languages also be based on these structures. From: json.org website In JSON, they take on these forms: An object is an unordered set of name/value pairs. An object begins with { (left brace) and ends with } (right brace). Each name is followed by : (colon) and the name/value pairs are separated by , (comma). An array is an ordered collection of values. An array begins with [ (left bracket) and ends with ] (right bracket). Values are separated by , (comma). A value can be a string in double quotes, or a number, or true or false or null, or an object or an array. These structures can be nested.

How do we extract the data in a readable format? Discovery Summit 2014 Data Collection How do we extract the data in a readable format? SAS has a JSON parser R has two packages called: RJSONIO: reads a JSON file HTTR: works with the web connection (See Appendix for example code)

Discovery Summit 2014 JMP exploration After formatting the data, we used JMP for visualization, exploration and mapping. Graph Builder: Flexible drag-and-drop data visualization builder Heat maps Bar graphs Bubble Plot: look at many variables at one time and animate Street Level Maps Background polygon maps Converted JSON New York City coordinates Converted a KML New York City map format to use as a background map Used a KML polygon importer JMP add-in Create custom icons Created custom icons using JMP add-in: Bubble Shape Editor KML polygon importer: file exchange

Questions we want to explore: Discovery Summit 2014 What are we looking to learn? Questions we want to explore: Where are the bikes, and how many are available? Where are the docks, and how many are empty? Who are the riders? Members or ad hoc riders? When do they ride? What are the popular neighbors? Station usage

Discovery Summit 2014 Let Us Show you Live JMP Demo

Discovery Summit 2014 Examples Examples of Output

Available Bikes or Docks Discovery Summit 2014 Available Bikes or Docks Available Bikes? Available Docks? Here we use custom Icons. Some thoughts. Time series: when will be the best time to get a bike or dock a bike? Need to determine timing pattern … days or hours ; Wkday, Wkend and Holiday

What time of day do they ride? Membership? Discovery Summit 2014 Ridership Who are our riders? What time of day do they ride? Membership? What are the popular neighborhoods by gender? Use Data Filter to drill down

Which day of the week has the greatest ridership? Discovery Summit 2014 day of the week Which day of the week has the greatest ridership? With Data Filter we can look at Gender break down, Membership etc.

Discovery Summit 2014 Neighborhood Usage See color patterns…. Just another view See the patterns…. Cluster to minimize interactions for modeling data When modeling look at neighborhoods, day of week, and time of day (add cluster information if applicable).

Application to look at station usage: Discovery Summit 2014 Station Usage Application to look at station usage:

If Citi Bike were to hire MnM consulting, here is what we would share: Discovery Summit 2014 Conclusions You can go from JSON format to formatted data to use with JMP exploration tools. We were able to visualize data to show Citi Bike where and how it could improve its business. If Citi Bike were to hire MnM consulting, here is what we would share: More males ride than females. Female riders prefer neighborhoods as opposed to financial areas. Sunday and Saturday mid-morning and afternoon are most popular. Certain neighborhoods and stations are more popular during certain times and days. How to manage bike stocking. Ridership in financial areas is consistent, but there are areas of improvement for Citi Bike.

Neighborhood station usage Discovery Summit 2014 WHAT we could show citi bike Time series: Neighborhood station usage Managing station needs: Which will need more bikes or more docks? Model: Understand the effects on available bikes and/or docks Look at additional data with Citi Bike data such as: Retail locations: Can we improve foot traffic? Is there an uptick around certain station locations? Restaurants and bars: Are we seeing more riders late at night? Subways: Is there more usage when subways are broken? Improve neighborhood connections and stations Citi Bike moves people …. People mover 

Other industries could use these same steps. For example: Discovery Summit 2014 Who else can we Help? Other industries could use these same steps. For example: Retail industry: Shopping patterns and geographical locations? Insurance: Where are the claims be filed? Types of claims? Geographical locations? Banking: Branch usage? Time of day and week? Most popular locations? Medical: Procedure patterns? Emergent care locations? Employers (and Employees): Another alternative to commute to work? Remember that Citi Bike is a company that provides a service to move people around a location. Stations must be accessible and easy to use. Location location location!

Discovery Summit 2014 THANK YOU *MnM Consulting Company thanks you for taking the time to view our presentation. Note: MnM is fictitious company name created for this presentation.

Sample Code for SAS and R Discovery Summit 2014 Appendix Sample Code for SAS and R

Discovery Summit 2014 SAS code to get and read JSON

Discovery Summit 2014 SAS Code

Discovery Summit 2014 R Code to get and read JSON file