PowerPoint Presentation

[Audio] Slide 1: Hello, we are Team Vikings here to present our exploratory data analysis project on two-wheeler electric vehicle data. Our goal was to analyze various aspects of vehicle performance and rider behavior to gain insights that are valuable for both customers and the company..

[Audio] Slide 2: We began by collecting and preprocessing the data, focusing on essential attributes like speed, acceleration, battery metrics, and GPS data..

[Audio] Slide 3: Through detailed data cleaning, dimensionality reduction, and visualization, we identified key patterns and relationships. This process was essential for ensuring reliable analysis and meaningful insights..

[Audio] Slide 4: We've carefully planned out our project and created essential documents to guide our work, including a Gantt chart to track our timeline.

[Audio] Slide 5: A Work Breakdown Structure to organize tasks, and a Software Requirements Specification to define our technical needs and goals.

[Audio] Slide 6: In our dataset of 297 features, we have carefully selected specific attributes to provide insights in three key areas: vehicle performance analysis, geographic analysis, and sharp turn and hard braking analysis. Let's dive into each area. By focusing on these key features, we're getting a clear, big-picture view of how well the vehicle performs, how users drive, and how different locations affect the ride. Our goal is that these insights will help drive new ideas for electric vehicle technology, making rides safer and more efficient for everyone. After that We concentrated on three key areas to gain a comprehensive understanding of vehicle performance. First, we analyzed battery efficiency and energy consumption in both Eco and Power Modes. Second, we conducted a geographic analysis of energy use across 63 cities to see how local factors impact efficiency. Finally, we studied hard braking and sharp turns to evaluate driving safety and its effect on battery performance. Each focus area provides actionable insights for both customers and the company..

[Audio] Slide 7: In this part of the analysis, we will examine various factors affecting vehicle performance, including the average speed for maximum range in both ECO and Sports modes, the speed at which the motor performs most efficiently, and additional aspects like the distance covered at specific speeds, the energy required for that distance, and how the State of Charge (SOC) is impacted at different speed levels..

[Audio] Slide 8: To analyze the average optimal speed, we divided the data into 5-minute time slots. In each slot, we calculated the total distance covered and total energy consumed to determine efficiency, defined as the total distance covered divided by total energy consumed. To avoid efficiency dominating the analysis (as intervals with low distance and low energy can show high efficiency but skew results), we created a combined score: Combined Score=0.5×Efficiency+0.5×Total Distance Covered\text = 0.5 \times \text + 0.5 \times \textCombined Score=0.5×Efficiency+0.5×Total Distance Covered. We then selected the 100 intervals with the highest combined scores and averaged their speeds to determine the final optimal speed..

[Audio] Slide 9: After running the above algorithm for each vehicle individually, we observed that the average optimal speed for all vehicles consistently fell within the range of 22 km/hr to 34 km/hr. After completing the final analysis and validating with the vehicle testing and performance team, we recommend that in ECO Mode, riders should aim to maintain a driving speed close to 28 km/hr to achieve optimal power consumption and maximize motor efficiency..

[Audio] Slide 10: Riders should aim to maintain a speed between 25-30 km/hr whenever possible, as there are a significant number of entries with speeds below 25 km/hr in the above plot. In ECO Mode, the top-rated speed is 45 km/hr, yet the plot shows numerous entries exceeding this speed. Customers should be advised to avoid crossing the top-rated speed, as it may negatively impact the overall range and battery life of the vehicle. It would be beneficial to explore the development of a new mode or motor optimized for the speed range of 38-42 km/hr, as the current optimal range of 25-30 km/hr may be limiting for certain users, who may require higher speeds for enhanced performance..

[Audio] Slide 11: In the above plot, we observe that many entries fall within the speed range of less than 45 km/hr. Since the motor performs most efficiently in Sports Mode at speeds between 55-60 km/hr, it is recommended that riders maintain speeds within this range when using Sports Mode. If the required speed is below 45 km/hr, riders should consider using ECO Mode to maximize range..

[Audio] Slide 12: The Purpose of the analysis is To reveal how energy consumption varies across regions, influenced by driving patterns and traffic conditions. And Key Insight from the analysis Comparing Eco Mode and Power Mode shows higher energy use in high-demand cities, indicating factors like traffic or challenging terrains..

[Audio] Slide 13: Higher Consumption in Pow Mode: Energy consumption in Pow Mode averages around 10 units, significantly higher than in Eco Mode, reflecting a performance-focused design that prioritizes power over efficiency. Top City - Angamali: Angamali again shows the highest consumption, suggesting consistent energy demands due to factors like terrain across both modes, with a more pronounced impact in Pow Mode. Uniformity Across Cities: Other cities, like New Delhi, Delhi, and Loni, display similar Pow Mode consumption, indicating a stable performance profile across urban settings regardless of location. Mode Efficiency Comparison: The increased consumption from Eco Mode to Pow Mode illustrates the trade-off between efficiency and performance, offering insight into the energy demands of each mode. Further Insights: Driving Preferences: Identifying which mode is commonly used in each city could reveal local priorities on performance vs. efficiency. City-Specific Optimization: Aligning mode usage with urban conditions may help improve energy efficiency based on local terrain and traffic..

[Audio] Slide 14: Higher Consumption in Pow Mode: Energy consumption in Pow Mode averages around 10 units, significantly higher than in Eco Mode, reflecting a performance-focused design that prioritizes power over efficiency. Top City - Angamali: Angamali again shows the highest consumption, suggesting consistent energy demands due to factors like terrain across both modes, with a more pronounced impact in Pow Mode. Uniformity Across Cities: Other cities, like New Delhi, Delhi, and Loni, display similar Pow Mode consumption, indicating a stable performance profile across urban settings regardless of location. Mode Efficiency Comparison: The increased consumption from Eco Mode to Pow Mode illustrates the trade-off between efficiency and performance, offering insight into the energy demands of each mode. Further Insights: Driving Preferences: Identifying which mode is commonly used in each city could reveal local priorities on performance vs. efficiency. City-Specific Optimization: Aligning mode usage with urban conditions may help improve energy efficiency based on local terrain and traffic..

[Audio] Slide 15: Suggested Vehicle Specification Modifications for Regional Optimization Throttle Sensitivity: Reduce throttle response in congested areas to prevent power surges, boosting efficiency. Power Delivery in High-Altitude Regions: Optimize power for uphill climbs in areas like Shimla, reducing battery strain. Dynamic Battery Management: Implement BMS adjustments to extend battery life in high-demand areas. Region-Based Driving Modes: Use GPS to auto-switch between performance and efficiency modes based on location. Tailored Maintenance Alerts: Provide predictive maintenance based on local driving conditions to ensure top performance..

[Audio] Slide 16: Hard Braking and Sharp Turn Detection: Our primary goal with this analysis is to improve rider safety, enhance vehicle durability, and provide valuable insights for maintenance and customer feedback. Vehicle impact includes increased wear on the brakes and tires, strain on the suspension, and reduced fuel efficiency. Rider impact, hard braking can lead to a loss of control, discomfort, and a higher risk of rear-end collisions. These factors highlight the need for minimizing hard braking events through better rider behavior and vehicle design..

[Audio] Slide 17: Trip ID The ECU dataset we're working with logs different data points from vehicles, but it doesn't really tell us when one trip ends and another begins. So, we've introduced a feature called 'tripID' to fix this, which will help us track vehicle trips more clearly, without this, it's difficult to analyze how the vehicles are actually being used. 'tripID' is a simple way to label each trip a vehicle takes with a unique number. A trip starts when the vehicle is turned on, and continues until the vehicle is turned off for more than five minutes. We chose five minutes as the cut-off to avoid counting short stops—like at a traffic light—as separate trips. Each vehicle has its own series of tripID, starting from 1..