Real Time Shovel Operator Feedback System

This project endeavoured to determine whether real-time knowledge-of-performance information would favourably influence the behaviour of a rope shovel operator. Previous work done by CMTE had established that overall machine performance was profoundly dependent on individual operators insofar as each shovel operator exhibits a unique operational technique resulting in varying machine performance. Since the truck/shovel operation is commonly employed on the production path of open cut mines and plays a significant role in determining mine's revenue stream, the benefits of higher productivity and decreased machine wear could be realised by positively modifying operator's behaviour to reduce performance variations.

Analysis of shovel field trial data conducted by CMTE determined that the highest stresses on the machine and largest variations in performance occur during the digging phase of the operating cycle. One condition identified to be prevalent amongst all operators was stalling of the swing machinery while digging. This "swing stall" condition occurs when the operator commands the shovel to swing whilst the dipper is engaged in the bank, that is, swinging while digging. The swing-during-dig condition produces unnecessarily high boom stresses and inhibits the smooth motion of the dipper through the bank, resulting in reduced productivity and increased machine wear. The shovel operator is unaware of the swing stall occurring during digging, as there are no visual or other cues present and, therefore, it is reasonable to assume that any change in operator's behaviour could be attributed to the feedback system.

A prototype system was developed capable of acquiring and processing shovel's electro-mechanical signals to calculate important shovel performance metrics. The metrics are used to present feedback information to an operator in an easily acceptable and understandable format. The Feedback System assists the operators in making correct decisions when unfavourable working conditions occur, i.e. swing stall, and thereby increases the operators' skill base to become more consistent with site-specific "operator best practice".

The corrective feedback solution uses visual indicators on a display screen to alert operators when a swing-during-dig event occurs and suggests a course of action that will relieve this condition. Extensive field-testing of the developed system at two different mine sites and across numerous operators consistently revealed that it is possible to:

1. Monitor and process machine electro-mechanical signals in real-time.
2. Assist operators with moment-by-moment decision-making processes.
3. Provide an intrinsically motivational training tool for operators to strive towards site-specific "best practice".
And, most importantly:
4. Influence operator's behaviour using real-time augmented visual feedback and thus improve overall shovel performance.
The shovel performance improvements are accomplished by reducing:
1. Cycle times.
2. Frequency of hoist stall events.
3. Component stresses.
4. Total energy consumption.