About the Customer
Chalo stands as India’s foremost bus transport technology company, recognized for its pivotal role in revolutionizing the public transportation experience. The company’s core mission revolves around enhancing bus travel, making it safer and more reliable for millions of commuters. Chalo accomplishes this through its innovative services, including live bus tracking and contactless payment solutions. The driving force behind Chalo’s vision is the acknowledgment that public transportation is a fundamental need for the vast majority of Indians, who often rely on buses due to cost constraints and worsening traffic congestion. The company’s commitment is to not only facilitate accessible and efficient bus travel but also to address the pressing issues surrounding private transportation.
Chalo confronted a series of intricate challenges that underscored the need for a transformative solution. At the forefront of these challenges was the pressing need for accurate and real-time Estimated Time of Arrival (ETA) predictions for various bus routes. Inaccurate ETA estimates had the potential to sow dissatisfaction among passengers and erode trust in the dependability of public transportation, a critical issue considering the reliance of many on these services. Furthermore, Chalo grappled with the complexities of handling data from diverse sources, such as S3 and PostgreSQL. The manual processing and transformation of this data for machine learning (ML) model training and deployment was resource-intensive and required automation. Lastly, the need for an adaptable solution was evident, as accuracy in ETA predictions was pivotal for Chalo’s competitive edge and its mission to make bus travel a preferred choice for the masses.
In response to Chalo’s challenges, Applied Cloud Computing Pvt Ltd (ACC) devised a comprehensive and forward-thinking solution that harnessed the capabilities of Amazon Web Services (AWS). The solution comprised several key components, each aimed at addressing specific pain points:
This SaaS based solution was initiated by configuring AWS Glue to establish an efficient data pipeline. This pipeline automated the extraction, transformation, and loading of data from diverse sources, including S3 and PostgreSQL, ensuring that Chalo consistently had access to clean, up-to-date data for ML model training and deployment.
To facilitate accurate ETA predictions, AWS Sagemaker Autopilot was utilized. This powerful tool was used to explore a range of ML models, including XGBoost, GBDT, Extremely Random Forest, and Neural Networks. These models were trained on historical data to create dependable predictions for bus arrival times.
The solution didn’t stop at model creation and best-performing ML model was deployed in real-time, enabling Chalo to furnish commuters with accurate ETA information for different bus routes. This deployment marked a significant advancement in the reliability of public transportation services.
Automation was at the heart of this SaaS based approach. An automated ML pipeline was implemented that periodically retrained the model with new data from S3 and PostgreSQL. This continuous improvement process ensured that ETA predictions grew progressively more accurate over time.
ACC collaborated closely with Chalo’s domain experts to set a high bar for accuracy. Together, they targeted a 90% accuracy rate in ETA predictions, which played a vital role in enhancing the overall travel experience and building trust among customers.
Results and Benefits
The partnership between Chalo and ACC bore fruit in the form of impressive results and a multitude of benefits. Foremost among these was the enhancement of the customer experience. The implementation of the ML-based ETA prediction system markedly improved the reliability of bus travel, offering commuters the assurance they needed to plan their journeys with confidence. This improvement translated into increased customer satisfaction and loyalty.
Operational efficiency received a significant boost through the automated data pipeline and ML model deployment. Manual effort and operational costs associated with data processing and schedule updates were reduced, leading to cost savings for Chalo.
Accuracy in ETA predictions were consistently achieved, with the ML models hitting the targeted 90% accuracy rate. This reduction in prediction errors-built customer trust and encouraged more individuals to opt for public transportation.
Chalo witnessed a surge in ridership as word spread about their enhanced services. This not only relieved the strain on private transportation but also contributed to a greener and less congested urban environment.
Furthermore, Chalo solidified its position as India’s premier bus transport technology company. This innovative SaaS based model solution and accurate ETA predictions set a new standard in the industry, providing Chalo with a competitive advantage that aligned seamlessly with their mission to improve public transportation and make travel better for everyone.
About Applied Cloud Computing (ACC)
ACC is a leading cloud implementation, system integration, and managed services company with a strong track record in accelerating cloud adoption and delivering tailored solutions. ACC specializes in supporting AWS workloads, including AWS CloudFormation, AWS Control Tower, AWS Lambda, and Amazon CloudFront. ACC holds key AWS competencies, including Digital Customer Experience, Financial Services Consulting, Microsoft Workloads Consulting, and Migration Consulting. We are an AWS Public Sector Partner and have attained Advanced Tier Services status, highlighting our expertise in delivering technology and consulting services within the AWS ecosystem.