.jpeg)
Case Study
Real-time Monitoring and Reporting for Government Digital Literacy Program
Challenge
A large-scale government digital literacy initiative, with 7.5 Crore enrollments, faced challenges in monitoring the integrity of exams conducted daily across numerous centers. The exams, running from 8 am to 8 pm, generated vast amounts of data, requiring a robust solution to prioritize surveillance and ensure fair testing.
They key challenges were:
-
Massive Data Volume: On average, 60,000 exams were conducted daily, with peaks of up to 1.5 Lakh exams, resulting in 42,00,000 data points being processed within 12 hours.
-
Resource Constraints: Limited human resources made it difficult to conduct surprise visits to all exam centers, necessitating a system to identify and prioritize high-risk centers.
-
Real-Time Monitoring Needs: A requirement for real-time data streaming and analysis to measure performance at multiple levels - classroom, proctor, exams - was required
Approach
Following were the key components of the system:
-
Real-Time Dashboard: A comprehensive dashboard was developed to provide real-time information on center performance, including metrics such as average candidate exam time and a credibility index based on AI-proctored assessments.
-
Prioritization Algorithm: Centers were evaluated against benchmarks, such as an average exam time of less than 5 minutes, to flag those with a high likelihood of suspicious activities. The dashboard used a color-coded alert system to help the flying squad prioritize visits.
-
Data Streaming and Reporting: Implemented Apache Kafka for efficient data streaming, Redshift for analytics, and Metabase/Jasper combination for reporting.
-
Additional services streamed real-time information on student registration and training progress, adding to the system's load.
Xponentium Impact
-
Increased Incident Detection: The red-flagged centers led to a 20% increase in detected incidents successfully, improving the overall integrity of the exam process.
-
Enhanced Decision-Making: The central team could track performance metrics more effectively, leading to better real-time decision-making and resource allocation.
