Header image

Tourism in Bangkok: Road to Recovery with Mapbox Movement

May 13, 2021 blog-post big-data data-analysis data-visualization geospatial geospatial-analytics location mapbox trends business government logistics real-estate transport

The COVID pandemic has greatly reduced outdoor activity worldwide, affecting businesses dependent on face-to-face activities, such as tourism. This is a major problem for countries like Thailand, where tourist receipts account for up to 14% of the country’s GDP (source). With an estimated 730,000 Thais unemployed in the last quarter of 2020, the government needed to display a sense of safety for visitors fearing the risk of infection (source). The government launched a range of solutions with varying levels of success, such as vaccination passports (source), quarantines in hotels (source) and even yachts (source).

To explore how tourists responded to these efforts, we used Mapbox Movement data to analyze activity in Bangkok, Thailand's tourism center. Using the mobility dataset’s granular location and time elements, we were able to see a slow increase in tourist activity due to government efforts, despite major tourist spots remaining closed.

Revitalization efforts increase tourist activity

We zoomed in on tourist sub-districts that showed the most increase in activity as of April 2021 versus the lowest point of lockdown last year on April 2020--Siam Square and Khaosan Road. Zooming in these sub-districts further, we found that areas with the highest activity were tourist areas.

For Siam Square (left), we found high activity in the mall complex and adjacent street, which had almost no activity during the lockdown. Besides this, a noticeable pattern in activity traced the BTS Sukhumvit Line, which connects directly to the Suvarnabhumi airport via the Phaya Thai station. These suggest an influx of tourists after relaxed restrictions, heading towards the crowd-favorite central shopping district. There was also increased activity near the floating Pratunam Market, albeit less concentrated.

For Khaosan Road (right), a similar increase was observed versus a lack of activity during lockdown. A well-known tourist transport hub, the street offers transport services to other Thailand destinations such as Phuket and Chiang Mai. This coincides with more relaxed government restrictions on 6 provincial tourist destinations in Thailand. With travel restrictions loosened, the data shows that mobility has recovered in these areas almost to the same level as last year.

Temples in the city remain closed

Next, we zoomed in on Bangkok’s other famous tourist and cultural attraction: the Buddhist temples. For two of Bangkok’s most famous temples, we found little surrounding activity--consistent with government mandates to keep them closed. We see the same for other major temples in Bangkok as well, such as Wat Benchamabophit, Wat Saket, and Wat Suthat.

Hotel activity rise looks promising for 2022

We infer occupancy in tourist accommodations by looking at activity near hotels. We were able to attribute activity for hotels given the data’s granularity, allowing activity to serve as a proxy for business performance. The proxy seems to reflect investor sentiment as well, as seen in comparison to the stock price of the top hotel group operating in Bangkok, the Erawan group. With an upward trend, the hotel industry appears to be on track to reach the 2022 government goal of full reopening despite a dip this last April from a 2-week no-gathering rule and a curfew on shopping and nightlife (source).

The granularity and time element of the Mapbox mobility data allows a unique perspective on current events, which can provide insight on policy impact and business performance. With this data, we have shown how mobility information can be used to better understand the state of tourism in Bangkok.

Interested in using mobility data for your use case? Contact us now or send a message to our friends at Mapbox!


Understanding poverty in the Philippines with artificial intelligence

Using publicly accessible datasets, we developed a low-cost AI model to estimate wealth for the entire Philippines and analyzed what geographical factors are associated with wealth levels.

A second look at PNP crime statistics: Taking population into consideration

A look at the number of index crimes per 10,000 residents for 15 Philippine cities.

Talks from our 'Humans + Machines' artificial intelligence conference

We're sharing the presentations delivered by the four speakers at our Feb. 19 artificial intelligence conference in BGC.