Bangalore as a city is infamous for a variety of things. Breweries in almost every neighborhood, delectable filter coffee that makes you smack your lips and insufferable traffic that makes you want to pull your hair out. Granted, a project on either of the first two would be far more palatable. Who wouldn’t enjoy seeing the craft behind different beverages and a city which does justice to them?
Unfortunately for the reader, what you will be seeing next is a detailed narrative on Bangalore’s unbearable traffic problem. I write from an autoethnographic standpoint, a victim of having spent the majority of my life in bumper-to-bumper traffic, and ambushed by a cacophony of the distasteful sounds these vehicles emit. If it’s any consolation, I only hope these visualizations will contextualize the traffic issue and not transport you to being in the midst of it. Nonetheless, do proceed to read with caution.
For some context, I was born and raised in Bangalore. As a Bangalorean, I can attest to how I’ve spent a fair share of my life in traffic. I often leave for plans well in advance to account for being stuck in different bottlenecks. I also spent countless days commuting on one particular stretch that evokes dread more than nostalgia- the outer ring road. The stretch of the outer ring road I’m referring to is from Hebbala to Silk Board Junction. it is one of the country’s most innovative regions, with multiple tech parks who contribute to a booming economy. With so many offices located on this stretch, it’s understandable that it also boasts residential areas, breweries, restaurants and more. The outer ring road though, is riddled with terrible traffic at every junction. It took me 2 hours to get to work, and 2 hours to get back home. Yep. I spent 4 hours a day commuting on the 500D bus route, where I’m sure I’ve probably left a piece of my soul and buckets of sweat behind.
I began this report with the intention to answer a couple of things:
1. How bad is traffic in Bangalore, especially on the outer ring road?
2. Is the traffic problem reflected across the board in different ways (eg: in congestion levels, vehicular speed etc.)?
3. What impact does the traffic have on the environment?
I also want to disclose that through some of my other courses I am working on a larger speculative design project to see what the outer ring road might look like in about 10-15 years. So, I do have questions of forecasting, urban planning and a bunch of other stuff in mind. This piece serves the larger project in giving me a visual of the current traffic to illustrate why we ought to look into this issue.
About the Data
For this report, I found the dataset on Kaggle and began exploring it before deciding to use it. The dataset itself was created using data from the Government of Karnataka. The provenance stated on the page is “This dataset is derived from open-source traffic data provided by the Government of Karnataka, recorded at various key intersections and roads in Bangalore. The data was further aggregated and merged from multiple sources to create a detailed dataset suitable for smart city and urban planning analysis.”
The dataset contains 8936 rows, spanning records taken from 2022 to 2024. Each record reflects the date that the entry was recorded, and corresponding details about it as follows:
Date: Eg = 1/22/2022
1. Location details such as the area of the city, specific road
2. Conditions such as the weather, if there was roadwork or construction
3. Traffic Activity: traffic volume, average vehicle speed, congestion level, travel time index, road capacity usage
4. Other factors affecting the road: count of incidents, parking usage, public transport usage, traffic signal compliance, number of pedestrians or cyclicsts
5. Environmental Impact
Visualizing the Data
In exploring the data, here is the final dashboard I created. I began by exploring the overall issue with traffic in Bangalore and whether it extends across the city. First, I chose to examine the congestion levels across all the areas of the dataset.
This bar chart depicts the congestion level across the 8 areas of the dataset (which also cover a substantial portion for Bangalore), illustrating that across 2022, 2023 and 2024, the average congestion level across every area is as low as 54.45% in Electronic City and as high as 93.99% in Koramangala. This is only the average level. This chart highlights that congestion levels in Bangalore are extremely high across areas. It covers the central business district of the city with areas such as M.G. Road, but also the outer ring road with areas like Hebbala and Whitefield, and even the southern portion of the city. Overall, we can conclude that traffic congestion is a problem across the city. The data certainly reflects that.
Following this, I wanted to dive into my focus on the outer ring road. I chose to filter out the road/intersection names and choose the 4 roads from the dataset which are a part of the outer ring road. The image below highlights the outer ring road in blue, with the 4 roads in the dataset that are a part of the outer ring road- Hebbal Flyover, Sarjapur Road, Marathahalli Bridge and Silk Board Junction. It should be noted that this doesn’t cover the entire outer ring road segment which is highlighted in blue, but the 4 roads do cover main junctions across this road.
Diving into the outer ring road segment, I wanted to illustrate the traffic issues here with how congested roads are, and how much time it takes people to commute while traveling on these roads. For this, I chose to examine 2 metrics- congestion levels on the outer ring road alongside vehicle speed. This is a visualization that especially took time to work through.
Iteration 1
Above is a recording of the bar chart which showcases the average congestion levels across the 4 roads, and the tooltip on hovering reveals the average speed of a vehicle on that road. For example, on Sarjapur Road, congestion levels are 93.85% and the average speed of a vehicle is 36.15 kilometers per hour (22.46mph). Whereas on Silk Board Junction, congestion levels are 53.70% and average speed is 43.33km/hr (26.92mph). Understandably, the relationship is such where average vehicle speed is lower where congestion levels are higher.
However, this bar chart doesn’t immediately signify that it is interactive and is meant to display vehicle speed alongside congestion levels. Through user feedback I gathered wherein I asked users to play with the chart and share their thoughts; no one went on to hover and look at the tooltips. Those who did, did not notice that average speed was one of the fields mentioned in the tooltip. I felt that a better visualization could convey the message I was attempting to put out there.
Iteration 2
Deciding that perhaps a static visualization would be better, I decided to create one wherein users can see the relationship between the two metrics at first glance. I looked into creating a dual axis bar chart.
The image above shows a dual axis bar chart, where congestion levels across the 4 roads are represented as bars and the average speed on each road is depicted as a continuous line. As opposed to an interactive version, this chart does provide both metrics on first glance, making it easier to comprehend. Users did prefer this over the interactive one, and the legend helped them determine which color refers to which metric.
However, users also expressed some confusion. One user said, “I’m not sure about the line. Is it, I mean, is the speed connected across all 4 roads? I don’t understand what it means.”
While I created this iteration, I found myself wondering the same and thought a clearer and simpler chart might work. Yes, there is a trend of speed decreasing as congestion increases, but the dual axis chart doesn’t best reflect that.
Iteration 3
For my final iteration I decided to cluster it and put two bar charts together. I created 2 bar charts within one, where users can see the congestion level on the upper panel and the average speed across all 4 roads on the lower panel.
The bar chart depicts the 4 roads on the x axis, and with 2 metrics in one, users can see the congestion level and the average speed for a specific road. The 2 graphs being stacked one on top of the other also showcases the inverse relationship between congestion levels and vehicle speed.
This chart received better feedback from users than the previous one. They understood the purpose of the visualization much quicker and immediately gauged the inverse relationship between the 2 metrics. Compared to the dual axis chart they said, “This is easier. I can compare them easily and see one go up and one go down.”
Environmental Impact
As I’m exploring traffic in Bangalore, I’m trying to situate it within the larger implications surrounding urban planning, city planning, the future of living in Bangalore etc. I’ve also lived in Delhi for a few years and experienced the terrible AQI, which makes me wonder if my throat is being smoked every time I inhale. While Bangalore isn’t at that level yet, pollution has I increased and I’ve noticed it over the years. I wanted to see If the data reflects the same, and how we can visualize the environmental impact of this traffic.
The image below depicts a scatter plot of the 4 outer ring roads, showcasing the traffic volume. The overall trend indicates that with higher traffic volume, the impact on the environment is higher.
As seen, Sarjapur Road with its highest traffic volume, has the most impact on the environment. The dataset here didn’t offer much clarity on the environmental impact metric other than a number. It doesn’t clarify the reasoning behind it or how it was calculated. Is the number reflective of carbon dioxide emissions? Is it nitrogen oxides? This does make it harder to work with and convey the meaning to the user. I guess before the user, as the creator of the chart as well, I can’t tell you what environmental impact means. I do wish I had more clarity on this. Apart from just showing the relationship between the traffic and environment, it would help to know exactly what number for environmental impact counts as high. It would help in weaving a better story, because we would have the exact information to convey.
Reflections
I think the aspect I’m struggling the most with is the aesthetic and theme of these visualizations. I have a story in mind and I am situating this in the midst of a larger speculative design project that I’m working on in other courses. It’s an overall question of what Bangalore, especially on the outer ring road might look like about 10-15 years in the future.
But the theme is definitely a struggle. I’m still thinking of a color palette for the entire project and I would want the visualizations to follow a similar theme. So as of now, it does feel disconnected and as though the choices were from a standpoint of colors that can just easily convey the severe congestion (red) and environmental aspects (green). Given that I’m someone who was born and raised in Bangalore, I suppose it’s also hard for me to condense the city I love so much into a simple color palette. It’s far too nuanced and complicated, and this gets in the way of allowing myself to think about a simplified palette.
User testing was especially helpful, and I would really love to expand my sample size and break it down across segments. For example, it would be really interesting to see how residents of Bangalore who live in areas close to the outer ring roads I chose perceive these visualizations.
I would also love to have access to the datasets directly from the Karnataka government, but I’m facing some issues in accessing the data directly, which is why I’m using the Kaggle dataset. It would especially help if I could get access to other roads on the segment of the outer ring road I’m interested in. I’d love the chance to explore that data deeply.
Additional data would also help me craft a better narrative. Currently I have the dashboard up and users can switch through tabs to view them, but that’s all. I think a larger scroll format with some storytelling peppered in here and there would do a better job. Perhaps a chance to map the data spatially and put together a story of traffic in better view would make for a richer visualization. Given that the larger scope is a speculative project, I’d definitely love to dabble a bit more with the data and also look at forecasting and trends. Bringing in the futuristic lens is something I want to work on with available data. A number of these factors came to mind once I completed my work. It’s something I intend to add on to.
I guess if there’s a reader out there who is interested, stay tuned. If it’s just the bots, I don’t suppose there’s much to say on the notes of sentiment for a city I hold so dear to my heart (in spite of repeated brake-ups in the traffic).