User Adaptive Score

The User Adaptive Score provides a dynamic and ongoing assessment of a user’s driving or riding behavior over time. Unlike trip-level scores that evaluate individual trips, the User AdaptiveScore is a rolling aggregate that reflects a user’s overall performance across multiple trips. This continuous evaluation helps in tracking long-term trends in driving or riding behavior, making it easier to identify improvements or areas that require further attention.

The user adaptive score is updated at the end of every trip and is computed using the following formula: User_Adaptive_Score = [(1 − α) × Current_User_Adaptive_Score ] + (α × Current_Trip_Score)

Where:

• Current User Adaptive Score: The user’s adaptive score from previous trips

• Current Trip Score: The score for the most recent trip

• Alpha (α): A weighting factor between 0 and 1 that determines the influence of the current trip score on the overall User Adaptive Score

GQL Definition: UserEngagementScore

GQL Path: user.engagement.scores.slice.name

query User($user_id: String) {
  user(user_id: $user_id) {
    engagement {
      scores {
	slice {
	  name # name of the score. Ex: OVERALL_SCORE, ATTENTION_SCORE, LEGAL_SCORE, etc
	  value # value of the score. Between [0, 100] for OVERALL_SCORE and between [0,1] for rest of scores ]
        }
      }
    }
  }
}

Sub Scores

The Dynamic Scoring Model features an adaptive overall score, denoted as OVERALL_SCORE, which represents a comprehensive assessment of a user's driving or riding behavior. In addition to this overarching score, there are specific sub-scores designed to highlight particular aspects of driving or riding behavior

Score	Comment
FOCUS_SCORE	Acknowledges periods of attentive driving or riding without active phone handling
CALL_WHILE_MOVING_SCORE	Identifies instances where drivers or riders avoid making calls while moving over a speed of 15 km/h
ATTENTION_SCORE	Acknowledges periods of attentive driving without distractions like hands free calling, handheld calling, phone handling or screen use
LEGAL_SCORE	Rewards adherence to speed limits
MFFS_SCORE	Variation of LEGAL_SCORE that excludes congested roads
HARSH_MOVEMENT_SCORE	Recognises sequences of smooth driving, emphasising gentle acceleration and braking
SMOOTH_SCORE	Similar to HARSH_MOVEMENT_SCORE but also takes into account harsh turning events

Example Wireframe

---

Additional helpful links:

• GQL UserEngagementScore definition