DataRobot Case Studies Predicting Carpark Capacity at Ascendas-Singbridge Using Machine Learning

Edit This Case Study Record

	Predicting Carpark Capacity at Ascendas-Singbridge Using Machine Learning DataRobot

Predicting Carpark Capacity at Ascendas-Singbridge Using Machine Learning

DataRobot

Technology Category	Analytics & Modeling - Predictive Analytics Analytics & Modeling - Machine Learning
Applicable Functions	Facility Management
Use Cases	Predictive Maintenance
Services	Data Science Services
Challenge	Ascendas-Singbridge Group (ASG), a leading sustainable urban and business space solutions provider in Asia, was facing a challenge with parking capacity at their properties. In densely populated cities like Singapore, parking capacity is a major issue. Despite having high-rise buildings with carparks or garages, parking capacity remained a challenge for both property managers and drivers. ASG wanted to forecast and predict parking lot capacity to optimize their parking services, improve the experience for visitors and drivers, and potentially increase revenue. They had previously used a different platform for model building, but it was costly and did not deliver the accurate predictions they needed. Read More
About Customer	Ascendas-Singbridge Group (ASG) is Asia’s leading sustainable urban and business space solutions provider, with Assets Under Management totaling more than $20 billion worldwide. Headquartered in Singapore, the Group has a presence across 11 countries in Asia, Australia, Europe, and the United States of America. They are focused on improving parking lot efficiency at their many properties around Singapore and throughout Asia. The company has more than $20 billion worth of Assets Under Management (AUM) in 28 cities across 11 countries, including Australia, China, India, Indonesia, and Singapore. Read More
Solution	ASG turned to DataRobot’s automated machine learning platform, specifically its Time Series features, to accurately forecast the capacity of their parking lots. They had previously used a different platform for model building, but it was costly and did not deliver the accurate predictions they needed. After a successful proof of concept with DataRobot, they found that the platform could generate more accurate predictions and was easier to use. The ultimate goal is to make carpark availability information accessible to drivers through an app. This would allow drivers to find which buildings around them have hourly parking spots available, providing a new revenue stream for ASG and a better parking experience for drivers. Read More Log in to view content
Contents

Technology Category

Analytics & Modeling - Predictive Analytics

Analytics & Modeling - Machine Learning

Applicable Functions

Facility Management

Use Cases

Predictive Maintenance

Services

Data Science Services

Challenge

Ascendas-Singbridge Group (ASG), a leading sustainable urban and business space solutions provider in Asia, was facing a challenge with parking capacity at their properties. In densely populated cities like Singapore, parking capacity is a major issue. Despite having high-rise buildings with carparks or garages, parking capacity remained a challenge for both property managers and drivers. ASG wanted to forecast and predict parking lot capacity to optimize their parking services, improve the experience for visitors and drivers, and potentially increase revenue. They had previously used a different platform for model building, but it was costly and did not deliver the accurate predictions they needed.

About Customer

Ascendas-Singbridge Group (ASG) is Asia’s leading sustainable urban and business space solutions provider, with Assets Under Management totaling more than $20 billion worldwide. Headquartered in Singapore, the Group has a presence across 11 countries in Asia, Australia, Europe, and the United States of America. They are focused on improving parking lot efficiency at their many properties around Singapore and throughout Asia. The company has more than $20 billion worth of Assets Under Management (AUM) in 28 cities across 11 countries, including Australia, China, India, Indonesia, and Singapore.

Solution

ASG turned to DataRobot’s automated machine learning platform, specifically its Time Series features, to accurately forecast the capacity of their parking lots. They had previously used a different platform for model building, but it was costly and did not deliver the accurate predictions they needed. After a successful proof of concept with DataRobot, they found that the platform could generate more accurate predictions and was easier to use. The ultimate goal is to make carpark availability information accessible to drivers through an app. This would allow drivers to find which buildings around them have hourly parking spots available, providing a new revenue stream for ASG and a better parking experience for drivers.

Impact #1	Ascendas has experienced a 20% increase in revenue by optimizing their carparks in the first eight months after the rollout of the project.
Impact #2	The Group IT team no longer had to manually engineer time series features such as moving averages, stationary and seasonality.
Impact #3	DataRobot significantly automated the data preparation and made the model building and deployment less time-consuming and easier to maintain.

Impact #1

Ascendas has experienced a 20% increase in revenue by optimizing their carparks in the first eight months after the rollout of the project.

Impact #2

The Group IT team no longer had to manually engineer time series features such as moving averages, stationary and seasonality.

Impact #3

DataRobot significantly automated the data preparation and made the model building and deployment less time-consuming and easier to maintain.

Benefit #1	20% increase in revenue by optimizing their carparks in the first eight months after the rollout of the project.
Benefit #2	Significant automation of data preparation, making model building and deployment less time-consuming and easier to maintain.
Benefit #3	Ability to scale the approach to new carparks without much additional effort.

Benefit #1

20% increase in revenue by optimizing their carparks in the first eight months after the rollout of the project.

Benefit #2

Significant automation of data preparation, making model building and deployment less time-consuming and easier to maintain.

Benefit #3

Ability to scale the approach to new carparks without much additional effort.

Download PDF Version

Overview

Predicting Carpark Capacity at Ascendas-Singbridge Using Machine Learning

Operational Impact

Quantitative Benefit