Our AI Technology Approach

At WideSense, we combine over 15 years of expertise in AI-powered control systems with purpose-built innovations designed specifically for the challenges of fleet electrification. Our technology leverages cutting-edge physics-based AI to transform how fleet operators plan, optimize, and execute EV operations.

The Limitations of Traditional Solutions

Traditional fleet electrification tools, whether based on physical simulations or AI, come with significant limitations:

Black-Box AI:

These rely on static, OEM-sourced data that fails to account for real-world variability like weather, traffic, or individual vehicle differences. The result? Predictions that may work on paper but fall short in reality.

Legacy Simulation:

While these models can handle large datasets, they require immense computational power, struggle with extrapolation, and often act as fragile, one-size-fits-all solutions that can’t adapt to nuanced fleet needs.

The result of these shortcomings is a reliance on guesswork. Operators are left estimating critical factors like how much energy an EV needs for a specific route on a cold day, how long it will take to charge, or whether a charging depot can handle the load—all without reliable tools to confirm their assumptions.

WideSense’s Physics-Based AI: A Smarter Solution

WideSense eliminates this guesswork with our proprietary physics-based AI, an innovation that combines the strengths of physical modeling with neural network adaptability. By integrating real-world data and scientific principles, we create digital twins that replicate the behavior of each vehicle and its components under real-world conditions.

Accurate Predictions
Our technology forecasts energy needs, charging times, and vehicle performance at a granular, vehicle-specific level, accounting for differences even between seemingly identical vehicles.

Dynamic Learning
By constantly adapting to new data, WideSense’s models stay relevant, ensuring optimal fleet performance across diverse operating conditions.

Optimized Operations
Using these insights, we design charging schedules, route assignments, and depot layouts that minimize costs, enhance efficiency, and ensure reliability.

Why Physics-Based AI is Superior

WideSense’s neural net-wrapped physics models stand apart by addressing the weaknesses of traditional approaches:

Smaller Footprint: Our models require less computational power, making real-time optimization practical and efficient.

Greater Precision: Unlike black-box AI, our models use physics-based architecture for better interpolation and extrapolation, reducing errors and outliers.

Composability: Models are modular, allowing targeted updates
(e.g., updating only the battery model when the EV battery is replaced)

Eliminating the Guesswork

WideSense takes the guesswork out of fleet operations. Instead of estimating how much energy an EV might need to complete a route on a cold day, how much power is required to charge a vehicle in time for its next route, or how long that charge will take, WideSense provides accurate forecasts at a granular, vehicle-specific level. This precision accounts for even the subtle differences between seemingly identical vehicles, ensuring that every operational decision is grounded in real-world data.

Predict, Learn, Optimize

Our technology leverages three core capabilities:

Predict: Using live digital twins, WideSense models vehicle behavior in real time, incorporating variables like energy efficiency, charging dynamics, traffic, and weather to provide actionable insights.

Learn: An automated learning system refines predictions by sourcing data from similar vehicles in similar contexts, continuously adapting to operational realities.

Optimize: Multi-objective optimization tools translate these insights into efficient, cost-effective operations, enabling fleets to perform reliably under even the most challenging conditions.

Data-Driven Design

WideSense integrates diverse inputs—vehicle telematics, traffic patterns, energy costs, weather data, and more—to provide a comprehensive operational perspective. These insights guide every decision, from fleet composition to charging schedules, delivering actionable decision support that improves performance, reduces costs, and enhances service reliability.

Examples of Factors We Model—and Why They Matter to Fleet Operators

WideSense’s ability to model critical operational factors at a superior level is what sets us apart. Our technology provides fleet operators with actionable insights that solve real-world challenges in ways traditional approaches cannot. Here are some examples of what we model and why they matter:

  • Vehicle-Specific Energy Needs

    What We Model: Energy consumption at the individual vehicle level, accounting for differences in battery health, powertrain efficiency, and real-world conditions like temperature, load, and terrain.

    Why It Matters: No two vehicles are identical, even within the same fleet. Accurately forecasting how much energy a specific EV needs for a route on a cold day improves scheduling flexibility, prevents low-SOC events, minimizes downtime, and ensures service reliability. For example, a fleet operator can confidently assign the right vehicle to a high-demand route without overcompensating with unnecessary charging buffers.

  • Charging Dynamics

    What We Model: Charging rates, charger efficiency, grid constraints, and optimal state-of-charge targets for individual vehicles.

    Why It Matters: WideSense calculates precisely how long it will take to charge a vehicle to the level needed for its next assignment, balancing charger availability and energy costs. This eliminates bottlenecks at depots, reduces energy expenses by avoiding peak tariffs, and enables more efficiency scheduling. For instance, operators can ensure that each EV scheduled for early morning service is ready on time without overloading depot chargers or incurring unnecessary costs.

  • Battery Health

    What We Model: Degradation patterns, charging behaviors, and lifecycle impacts on battery health.

    Why It Matters: Battery replacements are one of the largest cost drivers for EV fleets. By modeling how operational decisions affect battery longevity, WideSense helps operators adopt practices that extend battery life, such as avoiding deep discharges or unnecessary fast charging. Additionally, by understanding how batteries degrade over time, WideSense can adjust scheduling and work assignments according to EVs’ actual capacities. This reduces long-term costs and improves vehicle and service reliability.

A Proven Impact

WideSense’s technology doesn’t just simulate the future; it actively shapes it. By accurately forecasting energy and operational needs and optimizing charging layouts, vehicle assignments, and depot energy use, we empower fleets to achieve tangible results. Whether it’s minimizing downtime, reducing environmental impact, or cutting costs, WideSense ensures your fleet operates with precision and confidence.

With WideSense’s physics-based AI, you move beyond guesswork to a solution that creates optimized, reliable, and cost-effective fleet operations—designed to handle real-world complexities.

Get Started with WideSense

Request a free consultation with our team to learn how we can help with your next tender.