Keeping Waterways Clean With Machine Learning

Situated near the equator, Singapore has a typically tropical climate with abundant rainfall. It rains an average of 167 days a year, with an annual total rainfall of around 2,500 mm. During heavy rainfall, silt can be washed from exposed earth surfaces in construction sites into waterways. Good Earth Control Measures (ECM) implemented in construction sites prevent silty runoff from being washed into waterways, keeping them clean and beautiful.

PUB, Singapore’s national water agency, receives more than 1,500 drawing submissions per year seeking approval for construction works that require Earth Control Measures (ECM) to be implemented. This process requires time as well as an experienced eye to manually review the drawings and ensure that the ECMs are properly designed.

Together with PUB, Zühlke found a way to augment the existing workflow using machine learning. Machine learning algorithms can analyze the input plans drawing and verify their design against ECM rules for compliance. Such an augmentation could reduce the overall time necessary to spend on verifying these plans by handling the more standardized cases and only requesting feedback from PUB officers for cases with higher complexity.

Over a span of four months, the team built a corresponding end-to-end prototype incorporating multiple models of computer vision (object detection and instance segmentation) trained on past ECM submissions to recognize key design elements. Furthermore, optical character recognition was used to extract relevant text and a rule engine checked compliance based on the recognitions.

In this talk, we present the process of ECM checking and its challenges. We dig into the machine learning workflow we built to augment this process and show how the resulting prototype looks like. We will discuss the challenges with the data we faced and our learnings during the project. An overview of the next steps will conclude the talk.

Speakers:

Paola Bianchi
Data Scientist
Zühlke Engineering AG

Silvan Melchior
Lead Data Scientist
Zühlke Engineering AG