Skip to content

ACCA Strategic Professional · Advanced Performance Management · Data science and analytics

Halvik Manufacturing uses an AI model to recommend bonus payouts to plant managers based on sensor-derived output, scrap and downtime data. Managers complain they cannot understand how the model reaches its scores, and the highest-scoring plant uses older machines with less sensor coverage. Which concern is MOST significant for the reward system's effectiveness?

The most significant concern is lack of transparency combined with biased data capture, which undermines perceived fairness and so motivation. Managers cannot understand the scores and plants have unequal sensor coverage. Reward systems work only if employees see them as fair and controllable, so the algorithm needs explainability and adjustment.

  1. ALack of transparency and potential bias in data capture undermine perceived fairness and motivationCorrect
  2. BThe model is too cheap to be credible to managers
  3. CSensor data cannot be used in reward schemes under any circumstances
  4. DBonuses based on algorithms always remove the need for managerial judgement

Explanation

Opaque models and uneven data coverage across plants threaten perceived fairness, which weakens the motivational link between performance and reward. Cost is irrelevant to this issue. Sensor data can be used in reward schemes, and algorithms do not always remove the need for judgement; here judgement is needed to correct bias.

Did you get it right without looking?

One question tells you little. A timed set on Data science and analytics shows your real accuracy, how long you take and where you lose marks.

Practise Data science and analyticsFree diagnostic mock

More Data science and analytics questions