Absolutely, the sample size has to be statistically significant. I think there is even a production hand book for floricultue which deals with experimental design for comercal production. They predefine the sample size and everything, to make it easy for large producers.
My first internship was at Alex R Masson in Kansas. One of my jobs was doing the new grow testing, for the production planing VP Dr. Wolff. I got to grow all the new virietes that we where planning to grow in the future, first and define which method gave the best results. Taught me a whole new way to think about a grow. How much did that space cost. How much labor. I got to play with both S3 and S4 everything was based on a 6.5 inch pot. Spacing models for plants we did not know. It was what convinced me to do production scheduling automation, computer where new and everything was by hand and slide rulers just for floricultue calculation, as my master topic. Some plants donāt need an S3 and S4 was better. Everything was how many turns per year. We knew the fixed cost of ever square foot of the space. Everywhere. Coolers where to have a 98% utilization year round. The werehouse and post harvest space could never be idle. Saleable product 52 weeks a year. Post harvest schedule very effectively.
I have to check the basement. I think I have the book still. It is really old school with tables in the back. Computer are for launching rockets aged mindset and reader. I had to reverse calculate lots of the tables. I wanted to convert all these queueing algorithm to relational mathematical function of a nest set in a matrix and build them in to SQL databases. DB2 was brand new only a beta.
I wanted to use an ISO 8602 date function and regression fill of space based on real sales numbers.
First wrote these in PL1, assembly, then in SQL. All the ISO math at the time had to be written assembly at first. Later we figured out by accident that it was a rather finnight calculation either direction just build a two column table for the date as a serial number 20190102 and the production week 53 this year. Hah index full both serial date and week by serial date. and then you have a qurryable dataset! From anywhere I can find the union of a batch and calendar.
Select c.week, c.date from iso8602 c where c.date = serial_date(today,8); how many years do you plan on Growing? That is the number of row. You need 27 years of rows to calculate every possible date before the set repeaters. There is an ISO standard for finance dates. That the date my week is calculated on to define when years start. You need 9682 rows to create a prepatual growing calender for cannabis.
@dan, your mom can use this for creating virtual projects in Quickbooks. You use the generally a YYYYWK, batch code id(serial number) like cutting a check. Everything you enter is accounted back to the batch code. Backs doing the books a lot simpler and your Grow is tied directly to quickbooks. Microsoft Project and Gant optimation built in based on an S3 or S4 model now everyone can automate there grow scheduling, down to when to order what based on your ipm SOP.
Post harvest is also accounted the same way.
Now everyone recalculate your space optimization and you will see a big problem.
I looked at the public record for each legal state. I can tell you the good ops and the bad ops. I donāt know the metiod of grow but I can tell you one thing. The average floricultue business cost 23 cents per sqft week. A sealed room cost about 37 cents pre sqft week on average. High tech hardened rooms cost just under a dollor per square foot week. That is your fixed operating cost. Whether you choice to use the space or not. I donāt Rember the fixed cost on lab space and food handling. But the US department of commerce has these numbers. Use floricultue for the grow. Use food handling for a poultry processor, chicken and you will have average fixed costs.
I take the worst plant possible in a grow. The dreaded poinsettia. We can not have cost in grows above poinsettias. Hydroponic tomatos with full supplemental light cost less than poinsettia per sqft week. There something wrong in our grows.
You add everything else cost of money, plus plants and supplies. Labor. Break you labor down to Full Time Equvolents FTEās. Planting 500 plants on week 53 takes how many people how many hours. If you are in the US the stand farm week has 40 hours in an FTE. But there are 2088 hours in a farm year vs most other business have 2080 work hours in a years. Nursing, police and EMT are on a 21 day work schedule and screw everything up.
If you qualify as a farm in your stateās property taxation take full advantage of the federal farm to table laws. Thatās all the federal specialty farm forms. We donāt get much but it better than a poke in the eye and much better then the corporate rates. Raise some chickens for eggs and put it into your ipm SOP and you can almost always can get a farm designation, a deduction, and free eggs.
Sorry but this is a summary of my masters and I wanted give the good Professor @fdousty a giant thank you. I will correct the spelling and grammar later.
You know that a comercal grower in floricultue who has a doctorete in pholosophy is addressed as Professor as a common practice.
From the voices in my head
Ethan.