This study is an investigation into methods of developing ergonomic guidelines for automotive assembly tasks involving insertion of small parts. The study was conducted in four major parts: 1) a method of determining and evaluating subjective exertion limits was modified and tested, 2) a large dataset was collected from an industrial population in 10 simulated assembly line tasks, 3) a smaller dataset was collected from a student population to assess hand dominance effects, and 4) strength data obtained was compared with a strength prediction model to determine if the model could predict manual insertion forces. The traditional method of psychophysical data collection requires participants to extrapolate sensations from a relativity short session to judge if the task could be done for a much longer period. Maximum acceptable limits (MALs) are typically derived from having participants adjust a weight, resistance, or frequency to an acceptable level. The present study evaluated a relatively new method of collecting MAL data for simple, single-digit exertions where participants were asked to determine an MAL by self-adjusting and then regulating to maintain the exertion level. Results showed that MAL values obtained from a series of self-regulated exertions were independent of both analysis method and duration (5 minutes vs. 25 minutes) used for evaluation, and that the method was repeatable both within and between sessions. Ergonomic guidelines are often obtained from the strength capacity for a certain task, as it is important to ensure that workers possess sufficient strength to accomplish a task. As task demands increase, however, a larger percentage of a worker's strength capability in required, and other factors, such as performance and worker comfort, tend to be compromised. In this work, both strength capacity and subjective limits were obtained for a variety of simulated tasks to facilitate development of guidelines for the specific tasks. The relationship between these two measures (maximum force, acceptable force) was determined, and acceptable limits were found to be approximately 55% of population strength capacity, with correlations (R2) ranging from 0.40 to 0.60 depending on the task, suggesting the subjective limits and strength capacity are related in these tasks. Hand dominance was found to have a small (5%), but significant (p = 0.006) effect on strength capability, and no significant effect on subjective limit. Biomechanical strength prediction models can be used to assess loads placed on the human performing various tasks. One of the more popular models, Three-Dimensional Static Strength Prediction Program, is often used for heavy material handling tasks, such as lifting or pushing. The tasks studied presently, however, are manual insertions, requiring localized force application rather than whole body exertion. The prediction capabilities of this strength prediction model were compared with strength values obtained from the simulated assembly tasks. Results indicated that the model was not successful when predicting localized force, accounting for only 40% of the observed variance in strength (R2 = 0.4) / Master of Science