Microsoft KB Archive/74270

{| = Solver's Linearity Condition Error Message Returned =
 * width="100%"|

Last reviewed: November 2, 1994

Article ID: Q74270

SUMMARY
The following error message may be returned when the model is known to be linear:

The linearity condition is not satisfied. Turn off the Assume Linear Model check box and then run Solver again

MORE INFORMATION
This happens when your initial guess for the solution is significantly different from the optimal solution and the Assume Linear Model check box is checked. This error message can be ignored in most cases because Solver has reached a suitable solution that is very close to the one that would be reached with the Assume Linear Model check box unchecked.

Steps to Reproduce Problem
  Create a spreadsheet with the following data: A1: 10000    B1: 1.50        C1: =A1*B1 A2: 10000    B2: 1.75        C2: =A2*B2 C3: =SUM(C1:C2)  From the Formula menu, choose Solver. Enter $C$3 in the Set Cell text box. Select Min. Enter $A$1:$A$2 in the By Changing Cells text box. Choose the Add button. Enter $A$1:$A$2 in the Cell Reference text box. Click the Down Arrow button and select >=.</li> Enter 0 in the Constraint text box. Choose OK.</li> Choose the Options button.</li> Check the Assume Linear Model check box. Choose OK.</li> Choose the Solve button.</li></ol>

The following error message appears: The linearity condition is not satisfied. Turn off the Assume Linear Model check box and then run Solver again $C$3 will equal -1.1E-12, which is essentially zero. Turning off the Assume Linear Model option will result in the same values.

Workarounds

 * 1) Choose OK to keep the current solution. It may be accurate.
 * 2) Turn off the Assume Linear Model check box and then run Solver again.
 * 3) Set the variable cells ($A$1 and $A$2 in the above example) equal to 1 or some smaller value and then run Solver again.