A special option in the Safety calculation is the Target ΣMsf loading type. At first this type of calculation works just like a normal Safety calculation using the phi-c reduction method: the program reduces the strength parameters and deformations are introduced. However, when during the calculation the specified Target ΣMsf is passed in a step (step i) the program starts to check for a number of criteria based on stiffness and deformation to ensure that the stable value for ΣMsf is larger than the specified target value. When these criteria are met the calculation process returns to the calculation step directly prior to the step that passed the Target ΣMsf value (step i - 1) and then takes a new calculation step such that the calculation ends exactly at the desired Target ΣMsf value.

Used criteria to determine if the ΣMsf has a stable value and at which point the calculation goes to the model state of step i-1:

1. the CSP (Current Stiffness Parameter, a measure for stiffness and plasticity) in the calculation is smaller than the CSP value of the specified Target ΣMsf value which was just passed in step i:
   CSP_current < CSP_{step i} / 100
2. the deformations of the current step are ten times larger than during the step of the specified Target ΣMsf value which was just passed (step i)

This process can be seen when plotting the displacements of a node against ΣMsf. See the example below: first the calculation calculates past the desired Target ΣMsf value = 1.15 (point A with ΣMsf= 1.17) until the two mentioned criteria are met (point B). Then the calculation jumps back to step i-1 (point C) to calculate the final step (point D, with Target ΣMsf = 1.15).

Since the graph of a pre-calculation selected node only shows the results of each step with a line between these two points, the last part of the curve looks a bit unexpected: since the calculation just uses the results from step i-1 (point C) and then directly calculates to the final result (point D), the line is drawn from point B directly to point D, while the actual calculation follows the path of point C to point D.