Date: 2024-05-18 Page is: DBtxt003.php txt00017028
Initiative in Data Science
CoClear
CoClear Press Release //Monday 17 April re: Columbia’s Data Science Institute, came together for a Hackathon co-hosted by CoClear
Peter Burgess
|
Date: 2024-05-18 Page is: DBtxt003.php txt00017028 | |||||||||
Initiative in Data Science | |||||||||
|
Burgess COMMENTARY Peter Burgess | |||||||||
| CoClear Press Release //Monday 17 April On Friday 31st March, 18 volunteer from Columbia’s Data Science Institute, came together for a Hackathon co-hosted by CoClear – a sustainability consulting group and Spry Group – a software development company. Participants in the hackathon were taught how to use D3.js to build data driven dynamic web content making a large CDP data set (previously the Carbon Disclosure Project) coherent and revealing the data at several levels of detail. In a first of its kind initiative, CDP collected the largest publicly available set of supply chain data from suppliers and their corporate customers across the world. The initiative, titled, the 'Supply Chain Program', involved over 75 global organizations and 4005 responding suppliers. In the quest to reduce emissions, CDP invited its members to answer questions specifically around the life cycle of individual products. To understand how the raw product data could be analyzed to inform strategic decisions, CDP collaborated with CoClear, to provide greater insight into sector and product-level emissions and their impact. CoClear ran detailed analytics of the life cycle assessments (LCAs) of 546 products (170 in 2013, 185 in 2014 and 191 in 2015) The idea of creating an interactive visualization available on a public website was advanced as a method of communicating those insights with the companies that participated in the program. The site could also become an attraction to sustainbility-minded consumers who want to gain insight into the carbon performance of various products, brands, and industry sectors by visiting the website and learning about the life cycle analytics of products. The interactivity of the visualization offers a way for visitors to click on products and drill down to gain a deeper understanding of the product’s supply chain. The development team from Spry Group led the teaching program dividing the students into 4 groups, Dragon, Tiger, Eagle and Shark – spending the better part of the day hacking. During Stage 1 students chose what in their opinion was the best chart to visualize the data from a selection of the D3 charts. Each group’s choice was guided by how they wanted to present the data. The chart needed to effectively show product level data that covered, at the upper level, a range of regions, countries, corporations and products - to upstream, downstream, operational emissions as well as carbon intensity at the granular end of the spectrum. During Stage Stage 2 the hackathoners learnt how to make scaleable vector graphic charts and in Stage 3 they were taught how to make the charts interactive. With expert tutoring and guidance available at each stage the students worked intensely through the day finally breaking in the late afternoon to present their individual visualizations at the conference room in the Data Science Institute. No small challenge for a day’s work. Representatives from CDP and the DSI were invited to judge the presentations providing an opportunity for the students to showcase their work. Judge Yue Quie and Andrea Tenorio from CDP and DSI judge Ali Mehani a post-doctural student at the DSI, forming the judging panel. Team Eagle: Yue Ulysses Chang, Chenchao Zang, Hazel Bing Team Shark: Zeyu Ye, Ke Ma, Zhenyu Wang Team Tiger: Stephanie Doctor, Yuting An, Jieyu Yao Team Dragon: Janak A. Jain, WeiWei Wang, Siton Zhou Team Dragon’s submission was selected as the winner with Team Tiger a close second. In the opinion of the judges Team Dragon’s visualization gave the user the ability to drill down into the most granular aspects of the data to the per product basis. Team Tiger allowed for a sector by sector and corporation by corporation level optionality, but with a less granular outcome at the product level. Data visualization has been recognized as a tool that enables the communication of large data sets in a more persuasive manner than textual information (Pandey). Beyond running the hackathon to explore methods of communicating publicly disclosed product information the groups involved seek to analyze the benefits of visualizing and sharing that data. At the time of writing this press release no attempt is known of that visually represents global product LCA data. By creating the world’s first online accessible visualization of product LCAs stakeholders from major companies such as GM, Bloomberg and General Electric who have submitted product LCAs to CDP can be encouraged to visit that site and check out the footprints of the products they submitted and compare to others. By comparing products in each sector with similar products corporations are encouraged to become as competitive on environmental performance as they are on other market factors. Please contact us for further details at sally@coclear. | |||||||||