Going Cognitive with Manufacturing Ergonomics
The demands of manufacturing processing and design are changing rapidly. Production facilities must be able to reconfigure processes and adapt quickly to the changing production demands of the market in order to remain globally competitive and profitable. As a result, production employees are required to do more. But more complex procedures and automation can increase the cognitive load placed on the process operator.
The typical manufacturing process design focuses on traditional ergonomics. This approach may not be adequate for modern manufacturing processes because traditional ergonomics emphasizes reducing the operator’s physical fatigue and discomfort to improve throughput and reduce safety hazards. The modern approach to designing manufacturing processes must include cognitive evaluations for effective implementation of system ergonomics. Therefore, the next generation of system design for the modern manufacturing process should consider not only traditional physical ergonomics but cognitive ergonomics.
Cognitive ergonomics examines how work affects people relative to their attention distribution, decision-making, cognitive aspects of mental load, stress and human errors. This is important for manufacturing processes because cognitive load can have an immediate impact on operator performance by slowing task performance and increasing errors. Manufacturing processes are moving from requiring more physical strength and endurance to a greater need for problem-solving and reasoning skills. With the growing demand of these cognitive changes in manufacturing systems, practitioners must consider this new approach to the human system design.
New human system design approach
Mental workload builds from cognitive demands, which form from the interaction between operators and their assigned tasks. Mental workload is an important measurement because it provides awareness about where increased task demands could hinder human performance.
In an attempt to address the changing demands of the manufacturing industry, the author developed a strategy that presents a framework for assessing mental workload in manufacturing processes. There are five essential steps to the framework: Study the manufacturing system, identify the cognitive elements, model the process, measure the mental workload (MWL) and mitigate work overload.
In step one, an ergonomist or process improvement team studies the manufacturing process to gain knowledge and insight about the system. Next, in step two, the evaluator uses an applied cognitive task analysis to identify the cognitive task elements in the process. The hierarchical task analysis and applied cognitive task analysis collectively define the process steps and cognitive elements, which are employed as discrete events in modeling the process, which takes place in step three.
This study used the Improved Performance Research Integration Tool (IMPRINT) as the human performance modeling tool to predict mental workload. IMPRINT was developed by the Human Research and Engineering Directorate of the United States Army Research Laboratory to assess human-system function allocation, human performance and mental workload estimation.
The discrete events, with their estimated task times and multiple resource theory ratings for their associated mental resources, were the primary simulation inputs for this strategy. Multiple resource theory, as developed by Christopher Wickens, is a predictive model that supports understanding how well an operator performs while multitasking. According to multiple resource theory, when the human mind receives task demands (inputs), it can distribute its resources to handle these task demands either individually or collectively. Resources can come in various forms, including visual, auditory, cognitive, motor and speech.
Task demands that overlap leave the mind with fewer available resources. Multiple resource theory predicts that human performance will decline when multiple tasks require competing resources, which could decrease system safety and the effectiveness of your manufacturing process. IMPRINT applies a scale to the mental resources identified by multiple resource theory to assess the mental resource utilization for completing a process task’s discrete events. This step determines the multiple resource theory ratings for each discrete event identified in steps one and two.
With this data, IMPRINT measures MWL in step four, outputs the workload predictions and provides a workload profile for the task under analysis. If there is an MWL overload, ergonomists should use multiple resource theory to conduct an assessment to attribute mental resources that correlate to the mental overload. (If a mental overload task element is not measured, this terminates the process.)
In the event of a mental overload condition in step five, the ergonomics team should modify the manufacturing system based on multiple resource theory principles to mitigate the overload. This is done by decreasing the mental resources that are creating the overload and applying the multiple resource theory ratings again to compare the modified system resources to the baseline process evaluation.
Once again, the MWL is measured using IMPRINT. The process will recommence until an optimal MWL range is established. Once an optimal range is attained, the altered job element sequence can be tested and validated on the production floor.
Recommendations for practitioners
As global competition continues to grow, practitioners and engineers must continue to improve upon the human system design process for manufacturing. The proposed strategy provides a baseline to build upon for future work in process development of repetitive task processes.
Since the development of this strategy, the framework has been expanded to validate the simulation modeling tool in the manufacturing setting. This was done by conducting laboratory experiments to test the simulation predictions to other MWL measures with evaluation tools from the various MWL categories of subjective, physiological and performance. The strategy has proven to be capable and effective in predicting overload conditions for processes with inspection and supervisory control (automation) tasks. However, an MWL threshold for manufacturing needs to be systematically established for effective process design.