Transitioning NYC heating fuel away from oil

Posted on Feb 4, 2018

Business Case

As of 2018, there are over seven thousand oil burning furnaces still in operation across the NYC area. Each boiler has an estimated lifespan of 35 years from the time of installation. The end of this life-cycle would be an excellent time for a natural gas provider to encourage a switch from oil to their product. Between 2018 and 2022, roughly 2300 oil burning furnaces scheduled for decommissioning, replacing these units with natural gas burners is worth $51 million a year in sales to natural gas providers (estimated from 2017 prices).  However, such a conversion requires an upfront cost of equipment and structural renovations creating a barrier to entry that may push many property owners to opt to stay with oil heat, switching to a lighter oil type to remain compliant with NYC fuel mandates, but resisting the overall push to completely move to cleaner natural gas.

Natural gas providers Con Edison and National Grid can profit from investing resources into community outreach programs to work with these property owners and encourage making a switch to natural gas. To help these companies focus their efforts, this app aims to answer three key questions:

  1. Where are these boilers located?
  2. When are they being retired?
  3. What are the primary cost/engineering hurdles of switching these units to gas?

The data

NYC Dept of Housing and Development published "Oil Boilers Detailed Fuel Consumption and Building" NYC Open Data 1. It contains over 8000 listing of oil boilers still in operation within New York City's five boroughs, with 42 fields of information for each entry. For the scope of this Shiny application, the only following fields were utilized:

  1. Facility Address
  2. Borough
  3. Council District
  4. Location latitude  & longitude
  5. Building type
  6. Natural Gas Provider (Con Edison or National Grid)
  7. Estimated retirement date of boiler (assuming 35-year average useful life)
  8. Total Estimated Consumption - High Estimate (MMBTUs)
  9. Total Estimated Consumption - Low Estimate (MMBTU's)

From these vectors, the following columns were calculated and appended to the database:

  1. Average Fuel Consumption (MMBTU) - Average of High and Low Estimates
  2. Price of equivalent units of Gas - Calculated from Average Fuel consumption and price of natural gas as of July 2017 2.

Borough and district shapefiles were also obtained from NYC open-data to visualize the data on a map.

The app

The core of this app is an interactive map with three control widgets to subset the data of interest and three additional controls over the map visualization. A link to the application can be found here. A time slider subsets the data to time ranges of interest. The user uses drop-down selectors to filter the data by gas-provider and borough of interest. The user selects the granularity at which the data is presented. two layers of choropleths summarize the number of boilers and the amount of fuel these boilers are consuming a year; the first layer aggregates the information by Borough and the second layer further subsets the information by council district.

App with default 2018 - 2020 year selection and Borough level map display

 

Selecting the borough of Manhattan and the council district view focuses on the region and provides oil boiler data on a more granular level. Take away text updates with the appropriate year range, number of boilers and potential sales of natural gas should those boilers be replaced

 

 

The shade of orange coloring gives the user an intuitive estimate of the relative number of fuel burned by oil furnaces by Borough or district. Numerical information can be quickly observed by hovering the cursor over the borough or district. Between the time slider widget, and Borough/ District layer toggles, the user can form generalizations of the 'when' and 'where' of when oil boilers are expected to be phased out. By choosing the 'points' layer on the map layer control, the individual buildings are projected and are color-coded by building-type, hovering the cursor over each point yields the building address, building type, MMBTU a year, and total square footage. A 'take-away' summary is projected onto the map at all times and informs the user the number of oil furnaces due for replacement in the selected Borough and year range, as well a the potential value of gas sales replacing those oil units would yield.

The data is summarized in a histogram and bubble chart presented in the "Data" tab. Functionally and results of this feature is discussed in the following section. The "About" tab summarizes the app's objectives and links to references and the data source.

Findings

The "Data" tab summarizes the data in a histogram and bubble chart. The selection widgets that were used to subset the map display apply to these graphs as well, so the user can pass through the years, gas providers, and Boroughs of interest. For the images below we included data for the next 10 years (to 2028), both natural gas providers, and all of NYC.

 

The top histogram that charts total fuel burned by the oil burners that are scheduled for replacement in that year. Each bar is filled by the proportional amount consumed in each Borough. Two observations are clear from this representation; 1) Manhattan and Bronx are the major areas of oil fuel being consumed that can potentially be converted to natural gas fuel within the next 10 years. Referring back to the map view through districts, one can identify key neighborhoods where these aging oil furnaces are concentrated, such as Washington Heights in Manhattan and Riverdale area of Bronx.  2) The window to make these changes predominately occur before 2025 with a peak of conversions occurring in 2022.  This time window applies to all boroughs and both natural gas providers. If there are efforts to attract these potential customers to natural gas, they should ramp up quickly within the next year and definitely before 2020. Using the app the subset by Borough shows the same pattern in all areas so this trend affects both Con Edison and National Grid.

The bubble chart compares the 27 different building types listed in the data set to identify which categories of building house the oil furnaces. This information may be used to tailor a marketing strategy and help identify common engineering challenges that may be encountered in different building types.  The chart compares the total count, fuel consumed, and cumulative square footage of the  27 building types. In all 3 categories, elevator-apartment buildings far exceed all the other building types with walk-up apartments following in second. The linear correlation between the number of buildings and total fuel consumed suggests that the number of units is the primary factor determining how much fuel consumed within each building category. Since it is New York City, elevator buildings happen to be the most numerous.

Conclusions

While there are still nearly 8000 oil burning furnaces operating in New York City a large bulk of these aging burners are scheduled for replacement in the next 5 years. Converting these units to natural gas heat will result in significant earning for the area's natural gas providers. New York City has an incentive program to encourage these properties to switch away from oil when the time comes, they are not yet obligated to. Natural gas providers could use this app as a reference to prioritize on the neighborhoods and buildings with oil boilers that are due for replacement, increasing their chances of gaining future sales.  Manhattan and Bronx are the largest concentrations of these oil units, however, there are still over 600 furnaces scheduled for replacement in Brooklyn, Queens over the next 10 years.  Replacing these is still worth 15 million in natural gas sales.

 

references:

  1. https://data.cityofnewyork.us/Housing-Development/Oil-Boilers-Detailed-Fuel-Consumption-and-Building/jfzu-yy6n
  2. https://www.eia.gov/naturalgas/weekly/

About Author

Michael Chin

Michael is a Data Scientist with a strong background in data exploration and data analysis in physical science fields. 10+ years of experience in all aspects of research, from hypothesis formation to data collection, preparation, analysis, and communication.
View all posts by Michael Chin >

Related Articles

Leave a Comment

No comments found.

View Posts by Categories


Our Recent Popular Posts


View Posts by Tags

#python #trainwithnycdsa 2019 airbnb Alex Baransky alumni Alumni Interview Alumni Reviews Alumni Spotlight alumni story Alumnus API Application artist aws beautiful soup Best Bootcamp Best Data Science 2019 Best Data Science Bootcamp Best Data Science Bootcamp 2020 Best Ranked Big Data Book Launch Book-Signing bootcamp Bootcamp Alumni Bootcamp Prep Bundles California Cancer Research capstone Career Career Day citibike clustering Coding Course Demo Course Report D3.js data Data Analyst data science Data Science Academy Data Science Bootcamp Data science jobs Data Science Reviews Data Scientist Data Scientist Jobs data visualization Deep Learning Demo Day Discount dplyr employer networking feature engineering Finance Financial Data Science Flask gbm Get Hired ggplot2 googleVis Hadoop higgs boson Hiring hiring partner events Hiring Partners Industry Experts Instructor Blog Instructor Interview Job Job Placement Jobs Jon Krohn JP Morgan Chase Kaggle Kickstarter lasso regression Lead Data Scienctist Lead Data Scientist leaflet linear regression Logistic Regression machine learning Maps matplotlib Medical Research Meet the team meetup Networking neural network Neural networks New Courses nlp NYC NYC Data Science nyc data science academy NYC Open Data NYCDSA NYCDSA Alumni Online Online Bootcamp Online Training Open Data painter pandas Part-time Portfolio Development prediction Prework Programming PwC python Python Data Analysis python machine learning python scrapy python web scraping python webscraping Python Workshop R R language R Programming R Shiny r studio R Visualization R Workshop R-bloggers random forest Ranking recommendation recommendation system regression Remote remote data science bootcamp Scrapy scrapy visualization seaborn Selenium sentiment analysis Shiny Shiny Dashboard Spark Special Special Summer Sports statistics streaming Student Interview Student Showcase SVM Switchup Tableau team TensorFlow Testimonial tf-idf Top Data Science Bootcamp twitter visualization web scraping Weekend Course What to expect word cloud word2vec XGBoost yelp