Repeatability

Picture this. A farmer, let's call him Dale, runs his first load of corn in on a Tuesday morning. Early harvest, light traffic, your most experienced sampler behind the counter. He gets a 14.9 moisture call, number two grade, clean ticket. He's happy. Drives home, tells his wife they're finally getting somewhere.

Dale comes back Thursday afternoon. Lines are backed up. It's hot. The regular sampler went home sick and you've got a part-timer covering. Same corn. Same field. Same combine. Different person, different probe pattern, different call. Now it's 15.4 and there's a B.C.F.M dock he didn't see coming.

Dale doesn't say anything right then. He just takes the ticket. But he's doing the math on the way home, and that math doesn't feel right. And next year, when his neighbor tells him the elevator on the other side of the county is “friendlier” to work with, Dale's going to remember that Thursday afternoon. Not the breakdown, not the line, not the price ,  just the feeling that the grade depended on who was standing behind the counter.

That right there is a repeatability problem. And it lives at the heart of why continuous improvement tools either earn their keep or collect dust.

I've spent a long time walking grain and feed operations, and I've watched the same story play out more times than I'd like to count. A team gets fired up about CI. They do a workshop, fill out some forms, and put action items on a board. Things get better for a stretch. Then harvest hits. The board stops getting updated. By February, you're solving the same problems you solved in May, because nothing about the process made the fix permanent.

The gap usually isn't effort. It's that the tools never got used the same way twice.

CI tools, whether you're working with 5S, standard work, layered audits, Fish-bone problem solving, or your own homegrown version of "here's how we figure out what went wrong". They only pay off when they're used the same way, every time, by everybody. A problem-solving form that gets filled out one way by the operations manager and a completely different way by the night shift supervisor isn't a system. It's just a habit that some people have and some people don't. In a grain facility, that inconsistency shows up fast. Not in your CI metrics, in Dale's ticket and check.

Where Most Operations Get Stuck

Every CI program lives somewhere on a maturity curve. At the bottom is fire-drill improvement: something breaks bad enough to ignore, you call a meeting, agree to do something different, and six weeks later the same problem is back wearing a different hat.

A step up is the workshop model, you improve things when you schedule time for it. This is better. You've got structure, root causes get worked, action items get written down with names and dates. But if the tools from the workshop never become the tools people reach for on a random Tuesday, you've built a nice island of improvement surrounded by a sea of business as usual. Come back after harvest and count how many of those action items actually held.

Consistent tool use is where things start to shift. The problem-solving form looks the same whether it's the operations manager filling it out or someone in their third week on the job. The 5S walk happens the same way every time. Layered audits cover the same points on every shift. You start to see patterns, not just individual problems. The same root causes showing up again and again in different places. And your people start to trust the process because it doesn't move on them.

At the top is embedded CI, the point where it becomes how you work. Problem-solving doesn't happen at a quarterly event; it happens at the morning meeting, at the scale house, in the control room. When a better way gets found, the standard gets updated. Near-miss reporting is high because people trust that bringing something up leads to improvement, not a lecture.

Your CI program sits right in the middle of a two-sided equation. On one side you've got your own people; operators, maintenance crew, supervisors, who need a consistent way to surface problems, work through them, and make sure the fix holds. On the other side you've got the farmers bringing grain in and the processors, mills, and ethanol plants taking it out, both of whom experience the results of whether your improvement effort is actually working. When the tools get used consistently, the grading call is the grading call whether it's seven in the morning or seven at night. When they don't, Dale starts to wonder if the grade he gets is a function of his corn quality or a function of who happens to be working that day.

Get Off the Porch and Go Watch the Work

One of the best things a manager can do to build repeatable CI is also one of the most skipped: go watch how the work actually gets done. Not a tour. A real, structured observation at the point of work, the scale house, the pit, the dryer, the loadout spout. In CI circles they call it a Gemba walk. Gemba just means "the actual place."

The discipline is to observe first and fix later. When you see something that doesn't match the standard, fight the instinct to correct it on the spot. Ask questions that open a conversation , "walk me through how you do this step", not questions that put people on defense. The sampler who developed a slightly different probe pattern probably has a reason. Maybe the probe stand is awkward. Maybe the handle sticks in cold weather. That reason is where your real improvement lives.

For grain operations, the highest-value moments are the scale house during peak receiving hours, the dryer at startup and during lot transitions, and the control room at shift change; the moments when work is under the most pressure and most likely to drift from the standard.

Close every Gemba walk with a note about what you observed and what you're going to do about it. A walk that produces no follow-up is just supervision wearing a fancier name.

Find the Real Problem Before You Try to Fix It

Most grain operations are pretty good at fixing things. They're a lot less good at making sure they're fixing the right thing.

Ø  5 Whys is exactly what it sounds like, ask "why" five times. The trick is not stopping too early. Most teams quit at why number two or three, which drops them at "operator error" or "equipment failure." Those aren't root causes; they're the door to the root cause. Keep asking. Why did the operator make that call? Because the standard wasn't clear. Why wasn't it clear? Because nobody ever wrote down the probe pattern for that particular pit configuration. Now you're somewhere useful.

Ø  Is/Is Not Analysis is underused and underrated, especially for problems that come and go. Map out where and when the problem happens, and just as importantly, where and when it doesn't. If grading inconsistency happens on afternoon shift but not morning shift, that comparison will point you at the difference a lot faster than asking why in the abstract.

Ø  The A3 holds it all together: background, current condition, goal, root cause analysis, countermeasures, implementation plan, follow-up. The discipline of fitting it on one page forces clarity. And it creates a record of the thinking, not just the action, so when the same issue tries to come back six months later, you've got something to refer to.

None of these tools are complicated. The hard part isn't learning them. It's using them consistently when the pressure is on.

Build the Process Simple Enough to Use Under Pressure

You can't have repeatable CI if the tools are so complicated that nobody reaches for them when things get busy. And in a grain facility, things are always either getting busy or getting busier.

A repeatable CI program has forms simple enough to fill out standing up during a packed receiving day. A visual board that gets updated because it's genuinely easy to update. A clear rhythm daily, weekly, monthly so people know when improvement work happens. And one person accountable for whether the tools are actually getting used, not just whether the last event got scheduled.

If your CI tools only work right when your most experienced people are in the room, you don't have a process,  you've got a dependency problem. Here's a gut check worth running: if your best CI person takes a week off in October, does the improvement process quietly stop? If yes, you've got a training gap.

Expectations and Accountability

You can't expect people to use improvement tools consistently if you've never defined what consistent use looks like. Everyone on your team should be able to answer in plain language: what does a finished problem-solving form look like at this facility? What is the grading standard, and who owns making sure it's followed? What makes a near-miss worth reporting, and what happens after someone reports one?

That takes written expectations not a policy binder, but clear, plain-language standards posted where the work happens. And it takes accountability that's consistent  not hammering people for imperfect paperwork but refusing to let it become normal for the tools to get used six different ways depending on who's on shift.

Accountability in CI isn't about catching people doing something wrong. It's about protecting the work your team put in to find and fix problems in the first place.

What Dale Looks Like When It's Working

In an operation with genuine repeatability, Dale's Tuesday load and his Thursday load get the same treatment. Not because the same person is behind the counter, but because the probe pattern is written down, the grading standard is clear, new samplers are trained to the same method, and a supervisor runs a quick layered audit a couple times a week to make sure the standard is holding. If drift starts to creep in, if one sampler's calls start running half a point high , the data catches it before Dale does. Someone has a conversation, the standard gets reinforced, and the problem closes before it ever shows up on a ticket.

Dale still doesn't know what a 5 Whys is. All he knows is that his moisture and B.C.F.M. calls feel fair, his grades make sense when he explains them at the kitchen table, and the elevator treats him the same way regardless of what day it is or how long the line is.

That's what he tells his neighbor. And that neighbor tells two more people. That is how a repeatable operation grows its book of business one honest ticket at a time.

Go Look at the Numbers Too

Gemba walks and problem-solving tools tell you what you can see. Data tells you what you can't. The two together are how you build a genuinely repeatable operation.

The most useful data in a grain facility isn't complicated. For receiving, track moisture and grading trends by variety  and by sampler. That second one is exactly what would have caught Dale's problem early. For drying, track actual versus target moisture out and gas consumption per bushel. For loadout, track moisture and grade at the door versus what the contract called for, plus any rejections with specific detail.

The discipline is deciding before harvest what you're going to track, collecting it the same way every shift, and storing it somewhere you can actually sort and trend it. A clean spreadsheet updated daily beats a sophisticated system nobody maintains. When the data is clean, patterns jump out, the sampler whose calls run consistently higher than everyone else's on the same grain, the dryer whose gas consumption has been creeping up for three weeks.

Data without action is just record-keeping. Connect what the numbers show to a decision, then follow up to see if it worked. That follow-up is what turns a one-time fix into a permanent improvement and it's the step most operations skip.

A Practical Place to Start

If you're not sure where to begin, start at the scale house. Pull three months of receiving tickets and sort moisture calls by sampler on the same commodity. If the numbers cluster tightly regardless of who was working, your grading standard is holding. If one sampler runs consistently half a point high and another runs low, you've found your gap — and you've found it with data, before Dale does.

Write down, in plain language, what a correct probe pattern and grading call looks like at your facility. Show it to the people behind the counter and ask honestly whether it matches what actually happens. Close that gap. Then check it once a week until holding the standard becomes habit.

Try this for a season. When your operators stop asking "are we still doing that CI thing?" and start saying "let me grab the form"; and when Dale stops doing math on the way home and starts recommending you to his neighbor, you'll know Repeatability is doing exactly what it's supposed to: MAKING YOUR OPERATION A PLACE PEOPLE CAN COUNT ON, EVERY TRUCK, EVERY SHIFT, EVERY SEASON.

Regards, 

Grain Guy Fifty