MRT Closures: What is the impact?
Written & Analysis By: Wing-Yiu, xkjyeah, Kai Wei
LTA and SMRT recently announced the closure of 17 stations at East West Line(EWL) and 2 North-South Line(NSL) on Sunday, 10th December and Sunday, 17th December. My colleagues and I were interested in assessing the impact of the impending closure by looking at (a) number of passengers affected and (b) the additional travel time required as a result of the closure. We hope that this data will be useful to members of the public and possibly private transport operators.
Number of Passengers Affected
We counted the number of people tapping into the MRT every hour and took note of where they tapped in and out, then averaged the number of passengers travelling on the MRT on all Sundays in Oct 2016 and Feb 2017. Following that, we plotted the data in a heat map table to show the amount of traffic between MRT stations.
The heat map table above shows the origin station on the left axis, the destination at the top axis. The average number of passengers who tapped in at each origin station over a one-hour period and subsequently tapped out at the destination station is represented by the colour of the cell (the darker the cell, the greater the traffic between the two stations).
If we study the passenger traffic on the East-West Line (EWL) along the affected stretch(Tiong Bahru to Joo Koon), we first notice that a significant amount of traffic stays within the affected stretch itself; that is, a significant number of MRT passengers travel within the West region.
What else can you observe? As there are so many things you can pick out from the chart, we invite you to explore the visualization for yourself. To navigate the interactive visualisation, click here.
Additional Travel Time Required
Besides knowing which MRT stations will be filled with passengers, we would like to know how much additional travelling time will be required for these passengers, assuming that they make the trips during the shutdown. We know that LTA and SMRT will be providing rail replacement bus services. However, we reasonably foresee that many commuters are going to take an alternative route comprising a mix of public bus and unaffected MRT lines rather than just the parallel bus service.
We plotted the additional time** required in a heat map just like the one we plotted earlier, except now a darker colour means a significantly longer travel time.
Again, through a simple heat map table, we can identify which origin-destination MRT pairs that require passenger to travel longer. To navigate the interactive visualisation, click here.
What other interesting observations do you have? Share your observations in the comment box below.
And most importantly, remember to plan ahead for your trips on 10 and 17 December!
If you do enjoy this post, do give us a 👏!
*Data pre-dates the opening of both the Tuas extension and the eastern leg of the Downtown line.
**At GovTech’s Data Science Division, we have built a public transport network model to analyse the public transport system, using ez-link data. Using this model, we measured the additional time required to make the trip between MRT stations, under the following assumptions:
- Commuters will always take the fastest route to their destination (not always true!)
- Commuters do not take the rail replacement service (this is because we do not have data on how long buses take to drive from station to station along the rail replacement route)
- Commuters take public buses within 200 m radius
- If there are more direct buses between any pair of MRT stations, commuters will take those buses instead of taking the train.
Please also note that this may not be representative of the additional time required by trips that do not start and end at MRT stations.
