This would be from sampling the wash water repeatedly to determine after how much of product X or hours used is there no longer and PAAleft to adequately reduce the microbial load - The system automatically doses PAA to maintain it's range, so it does not get to the point of there being no PAA left. There will always be PAA unless the supply runs out.
The PAA will last a lot longer than chlorine, however, the agitation factor will greatly reduce the length of time you get for each washer. Is there make up water being added constantly and if so, is it being dosed to 120 ppm? - Yes, there is make up water being added and it is dosed proportionally. The water is also recirculated i.e when displaced it funnels back into the re-circulating tank and is pumped back out again.
So, hopefully you've been checking the ppm since inception and you simply need to input that against the #KG of product that was washed and do the math
If you have not been checking the ppm faithfully, that's where you should start.......once an hour for at least a week.........then do the math against what was washed. Fortunately the system does come with an internal datalogger that's set to log ppm every minute. When you say do the math, I presume you were thinking of a system that doses once and then stops and to do the math of figuring at what point the PAA concentration falls below specification? As mentioned, this system automatically keeps dosing to maintain the ppm level.
Obviously this will vary depending on the microbial load of the produce at any given time, but you should account for worst case scenario when deciding. I know organizations are really concerned about the cost of potable water.
In poultry chill tanks, I have sampled the water once per hour for a month for:
e coli
salmonella
level of PAA
Once we were able to prove that with the maximum amount of carcasses in the tank, and zero e coli and salmonella, the wash water was used for more than 1 day
So this really depends on what you find in the water
Good luck (this is going to take some time to do properly)
The other variable we'll also need to consider is 'taint'. Some fruits, vegetables or salads can leave behind a colour, taste or smell that would affect the next incoming material. A good starting heuristic then would be to change the water between different groups.