I'm still in the progress of figuring out how to apply it to myself. I'm currently using some lifts/weaknesses in a concurrent framework, and some lifts more GBR-style. I want to migrate completely to concurrent. However, I still need to take some time to develop a proper algorithm that works for me. Currently, I don't have a proper algorithm yet, but it will be something along the lines of:
1) Assess your weaknesses and lifts you want to improve
2) Pick 3 to 6 lifts that carryover to the weaknesses and/or lift you want to improve. Be creative (think partial ROM, bands, unconventional lifts, etc.). The more advanced you are, the more lifts you need to pick.
3) Pick secondary exercises to fill up the gaps and increase volume a bit (e.g. side raises, abs, curls). These may also vary, no need to keep them constant.
4) Rotate (or randomize for that matter) the exercises each and every workout.
5) Focus on progress. Be creative here as well. Use cheats, negatives, different rep-ranges, different RPE, explosive, controlled, whatever works best for you and the lift.
6) Keep doing this until either you develop new weaknesses or you want to improve another lift (or until you want to become a cardio bunny).
This protocol allows to use optimal volume (30-70 per muscle group per workout) and optimal frequency (e.g. 2 times a week in an upper/lower format). You can also think about applying some form of DUP (heavy and moderate days).
Some minor notes that may help as well:
I seem to respond better to cheats where muscles are more involved to just holding on to the weight/bounce instead of solely contract. Also, I note that heavy partials translate surprisingly well to full ROM lifts. For example: my RDL increased significantly once I started doing rack pulls.