User Tools

Site Tools







Tech Start Ups

Safety Requirements


WGA GHAI Updates



Here are a list of topics and questions that harvest automation startups focused on apples should think about as they build their overall strategic plan and product roadmap and set development priorities.

Market sizing

How big is the market?

Provide information from industry groups or individual growers with source of data provided.

1) How big is the market in terms of apples (count) and GDP ($ sales)? This is important because it helps set the market size for investors and helps determine the cost structure for the startup.

 a) # of apples picked/harvested per year 
    i) By region (state/country)
    ii) By apple type (red delicious, honey crisp, etc.)
 b) Total $ value of apples 
    i) Retail
    ii) Food service 
 c) Addressable market is the amount of harvest labor performed to pick all those apples 

2) Grower metric - this is important because it sets the target for the product development roadmap. By knowing this number, startups have a reality check # to rein in over-engineered solutions that will not work within grower economics.

 a) Harvest labor per bin is the key grower metric 
 b) This should be the focus of grower discovery calls - get to that number for the type of apples they grow (different breeds have different growing and harvest techniques due to genetics differences - some are more fragile or require different picking methodologies when automating) 

3) Horticultural practices - these are important because they impact the design and development efforts for startups and help provide a full picture of the harvest and harvest-related activities. All of the activities should be looked at to help maximize the impact of harvest automation.

 a) How does the apple grow and what practices can impact the harvest process? 
    i) Pruning 
       a) When - post-harvest or during dormant season 
    ii) Thinning 
        a) Chemical thinning
        b) Green fruit thinning 
 b) Growing system 

1. Post-harvest 2. Dormant ii. Thinning 1. Chemical 2. Green fruit iii. Growing systems iv. Growth – subjects to know 1. Doubles – how to pick 2. Blooms – 5 per spot a. King bloom – main bloom b. Side blooms – supporting blooms c. Bin dimensions and management practices

3. Key requirements a. When is the apple ready to be picked? b. Understanding the damage requirements i. Culls ii. Grading standards c. Ripeness i. Tied to planned consumption timeframe i.e. less ripeness for apples with distant consumption timeframes d. Is there a requirement that doubles be picked? e. Is there a requirement around how many apples can drop to the ground during harvest? f. Stem clipping i. Is it a first-generation requirement? ii. Can it be optional? For example, by including the ability for a person to stem clip alongside the harvester. iii. Data 1. Honey crisp – clip every stem a. Most valuable – so less damage b. Easiest to damage c. Woodiest, knobbiest stems 2. Fuji – some clip, some do not a. SuperFresh – wants to stem clip everything iv. Stem clip is clipping the stem off the apple parallel to the top of the apple bowl – the woody knob is then short enough so it doesn’t damage anything else v. 3 choices 1. Automate 2. Have people with available stem clip – optional if grower wants 3. Don’t stem clip at harvest vi. H2A labor is priced with or without stem clip (higher with) g. Bin Management i. Range of bin sizes to be supported ii. How will bins be supplied and received from the harvester? iii. Sizing differences 1. Bins are not the same size 2. Runners are not the same size iv. Have the apple industry gather data around the range of sizes 1. Fill it from the middle out v. Packing houses have to worry about it when they’re storing and transporting – different sizes can be a problem h. Imaging systems in packing houses i. They can do sub-dermal images and cull ii. Dan – harvest does not need to do that level of imaging 1. Don’t need to do this with robots – humans can’t see sub-dermal iii. Opportunity to do the cull at the field and once harvested just drop it on the ground 1. Should be on the roadmap but should not be version 1.0 product requirement iv. Different kind of imaging problem on the harvest equipment 1. Have to deal with lighting variation 2. Imaging systems in packing houses v. World leader in packing house imaging 1. Compaq – world leader in sorting system for apples and stone fruits 2. European competitors – Grefa and others 3. Compaq got it right for the apple industry 4. Key guys a. Warehouse manager b. Line manager 5. The visioning systems that deliver 4M apples/day could only do 1M day without the visioning systems vi. Pack houses have more technology 4. Manual harvest process a. What do they teach the manual harvesters and how does it improve their picking capabilities? b. How is the quality of the pick judged? i. For manual harvest – usually by the crew manager or the orchard manager ii. For automated harvest – use the same evaluation metric as for manual harvest 5. Manual harvest economics a. Common economic metric for apple harvest = $x/bin picked i. Many WA apple farmers are in the $20-$25/bin range ii. Which generally works out to $.02-$.03/apple (1,000 apples per bin) iii. See Dan Steere’s email with H2A data b. Make sure that the economics of manual harvest are fully loaded i. What costs are there that are required to be paid for manual harvest? 1. Example – H2A housing costs need to be included on a per harvest labor resource basis so that they are fully loaded for any portfolio mix (see below) 2. Growers need to include the H2A costs (and if only used for harvest labor than the entire H2A cost has to go against the harvest labor total cost ii. Get the grower to work through the math related to regulatory costs 1. This will be state dependent and depend on what is current legislation and what is phased in + what will be phased in over the next few years (see minimum wage in CA as example) 2. So for that state you should know the current and future costs for the following components: a. Minimum wage i. Know what the current minimum wage is and know what is happening to it over years b. OT – CA rules i. Phases in at over 8 hrs/day ii. Also phases in at over 40 hrs/week c. Multiplier i. 1.5x/during week OT ii. 1.5x/Saturday iii. 2x/Saturday d. Breaks i. Every 2 hours (CA) ii. Lunch break every 4 hours iii. Understand the economics dynamic between piece rate picking payment and minimum wage 1. In short, farm workers get the higher of minimum wage or piece rate at all parts of the season 2. So if there is less volume to pick during certain portions of harvest season as the harvest rate starts to ramp, minimum wage will act as a floor (most likely toward the early and later part of the harvest season) 6. Manual harvest and automation working together a. Is there a viable model if the automation solution can only pick 50% or some other % of apples? b. Confirm with the grower that they view all harvest options as a portfolio to modify as mechanical harvest solutions improve and earn an increased share of harvest c. Confirm with grower that they have estimated the change in labor cost estimated for the next 3-5 years i. That is the true cost that startups are trying to move towards so they can get a larger share of the harvest portfolio

apples.txt · Last modified: 2022/10/29 23:59 by