🚀 Access the SPADE Web Tool

Turn TSDD Data Into Clarity with AI-Powered SPADE

Trained AI Models

Physics-Informed AI

Learns from full slope curve patterns

Handles noisy sensor input

Measurement Profiles

Reconstructed TSDD Profiles

Builds complete slope profiles

Converts slope measurements into full deflection curves

Quality Check

Automated QC

Applies a 3-layer consistency framework to detect anomalies and inconsistencies in TSDD slope data.

Full Slope and Deflection Profiles Powered by SPADE

SPADE reconstructs complete slope profiles from TSDD measurements and integrates them to predict smooth surface deflections.

Flagging Inconsistencies with SPADE’s QC Engine (Example 1)

Residual analysis compares SPADE’s slope to measured values at each sensor. In this figure, sensor s_130 is flagged because the residual (difference) exceeds the allowable threshold.

Flagging Inconsistencies with SPADE’s QC Engine (Example 2)

The Envelope check validates each sensor’s slope against its expected physical range (between P1–P99 percentiles). SPADE flags any slope that falls outside these limits, as shown for sensor s_-200 in this example.

Flagging Inconsistencies with SPADE’s QC Engine (Example 3)

This figure shows multiple flags based on SPADE’s 3-layer consistency framework—Envelope Check, Residual Analysis, and Goodness of Fit (R²)—using pre-defined thresholds for each test.

FAQs

Answers to questions you might have about SPADE

What is SPADE and what does it do?

SPADE (Slope and Physics-informed Analytics for Deflection Estimation) is an AI-powered tool that reconstructs full slope profiles from individual TSDD deflection slope measurements and calculates full-surface deflection curves. It enables users to perform quality consistency checks on measured slope values, ensuring more reliable data for analysis and decision-making.

Who can benefit from using SPADE?

SPADE is built for TSDD data providers, consultants, and transportation agencies (DOTs) aiming to enhance deflection analysis, perform automated slope quality checks, and extract deeper insights from pavement deflection data. Vendors can also integrate SPADE into their workflow before data delivery, improving quality and value for end users.

What makes SPADE different ?

Unlike traditional deflection methods that rely solely on raw sensor readings, SPADE uses a physics-informed machine learning model trained on slope profiles. It reconstructs complete slope curves—even when some sensors show anomalies—and performs automated consistency checks to flag inconsistencies in measured deflection slopes.

How accurate is SPADE?

SPADE has been benchmarked using both field and synthetic datasets and shows strong agreement with full-reference profiles. Its physics-informed model reconstructs a complete deflection profile from individual TSDD slope measurements—even when some sensors show anomalies. While SPADE can mitigate inconsistencies in the data, the final accuracy still depends on the quality and reliability of the measured slopes.

How do I get access to SPADE?

To access SPADE, simply reach out to RoadVision Analytics through the Contact page. SPADE is currently available via partner projects or tailored access under NDA.

Unlock the full potential of your TSDD data with SPADE

SPADE delivers physics-informed slope and deflection analysis to elevate your TSDD data quality.

For any inquiries, please contact info@roadvisionanalytics.com.