I think you are looking at it from the wrong angle. Write the programme as a logical sequence, and the resourcing will come from that.
Once you have that, look at what tasks would benefit from more labour, and add them in IF there is an overall benefit. Similarly, if 2 tasks can be carried out at the same time then programme them as such.

If you try to programme to a constraint on labour, chances are that the project will not go as planned. More important to write an achievable programme then optimise where it has benefit, than aim for a theoretical target.

also bear in mind that risk should be factored in to the programme, and this will upset labour levels, If you have 12 units available, dont go over 8 or 10, as overruns will place strains on resources you havent got.
this also allows for snagging during the job, which will also consume manpower units. backend works will also more than likely end up being late, and need more resource than planned.

Concentrate on a writing a cohesive, logical programme, and th erest will come from that