AI by department · operations
AI in operations: the work keeps moving, even when the planner is on holiday
For operations, planning, logistics and work preparation: the departments where orders come in by email, PDF and portal, and where the knowledge sits in a few heads. AI puts orders and requests into your ERP, TMS or WMS, makes a planning proposal, prepares the work and writes the report. The planner decides.
- Van Berkel: 192 hours a month back
- Connects to ERP, TMS and WMS
- The planner decides
- InRequest by email or formOver 100 a day
- AIRead and filteredWhat is it, for whom, how urgent?
- AIDraft in your own systemAll fields filled in
- HumanAn employee approvesExceptions go to a human
- DoneProcessedReady for planning
Van Berkel: 192 hours a month back
Worked with us
01 / What it is
What does AI do in operations?
AI in operations takes over the work between 'something comes in' and 'it's in the system and scheduled'. Orders, requests and waybills are read and entered into your ERP, TMS or WMS. A planning proposal is made based on all the data. Work preparation and estimating get a first draft from earlier projects. And the report writes itself. The planner and the work preparation engineer decide.
Operations lends itself well to AI. The volume is high, the work comes back every day and mistakes are expensive: a wrongly retyped order costs a trip, a delivery or a customer. And there's something else. The knowledge sits in a few heads. The planner who knows which driver doesn't want to go to Germany. The work preparation engineer who knows that customer X always forgets to mention the size. If they're on holiday, or retired five years from now, things grind to a halt.
AI doesn't solve that by replacing the planner. It solves it by capturing their knowledge, taking away the retyping and giving them a proposal they can work from quickly. They decide faster, and their colleague can too.
01Orders and requests
Email, PDF, Excel or portal: straight into ERP, TMS or WMS as a draft.
02Planning
A proposal based on all the data. The planner decides.
03Work preparation and estimating
A first draft from earlier projects and the price list.
04Knowledge retention
What's in people's heads is in the system too.
02 / Sound familiar?
The work keeps moving. As long as everyone's there.
Operations runs on craftsmanship and retyping. You want to keep the first, not the second.
Retyping
“Orders come in by email, PDF and Excel. Two colleagues do nothing but retype all day.”
The planner wants a holiday
“They haven't had two weeks off in a row for three years. Nobody else understands the schedule.”
The customer's portal
“Every big customer has its own portal. Every morning: log in, download, retype.”
The estimate
“A quote for custom work takes two days. Only our senior work preparation engineer knows how to do it.”
Status queries
“Where is my delivery? Twenty times a day, and three systems for every answer.”
The weekly report
“An hour of exporting and pasting on Friday. Nobody reads past the first page.”
03 / Case
Van Berkel Professionals: 192 hours a month back on request processing.
At Van Berkel Professionals, requests came in by email and through forms. Over 100 a day. Every request was read by hand and retyped into their own system. Work that had to be done, but that nobody got anything out of.
Now the incoming flow is automatically read, filtered and entered into their own systems. An employee approves every request before it moves on. The result: 192 hours a month back, about two FTEs of capacity that now goes to other work.
What makes processes like this so suitable: the volume is high, the input is digital and you know exactly when it's right. Those are the three things we look for in every AI Quick Scan.
01Volume
Over 100 requests a day.
02What AI does
Reading, filtering and entering into their own systems.
03Where the human is
An employee approves every request.
04Result
192 hours a month, about 2 FTEs.
04 / Before and after
A week in operations, before and after AI.
Six tasks as we build them in operations, planning and work preparation. The requests come with a client figure; what the rest delivers for you, we calculate with your own volumes.
- Van Berkel, requests: 192 hours a month
An order or request by email
Before
Open the PDF, retype items and quantities into the ERP, check the address.
With AI
The order or request is in the system as a draft with all the details. The order desk approves, exceptions are set apart.
The weekly schedule
Before
The planner spends a day puzzling in Excel and in their head.
With AI
A planning proposal based on orders, capacity, agreements and preferences. The planner moves things around and decides.
An estimate for custom work
Before
Two days of calculating, calling suppliers, searching for comparable projects.
With AI
A first estimate from earlier projects, bills of materials and the price list. The work preparation engineer checks the assumptions.
Work preparation
Before
Drawing up job sheets, material lists and instructions from the order by hand.
With AI
Job sheet and material list as a draft from the order and similar jobs. The work preparation engineer fills in the rest.
- Indication: 3 to 5 min per query
Where is my delivery?
Before
Open the TMS, call the carrier, email the customer back. Twenty times a day.
With AI
The reply is ready with the current status from the TMS. The employee sends it.
The weekly report
Before
Export, paste, format. An hour on Friday.
With AI
Delivery reliability, lead times and deviations in one overview, with an explanation of what stands out.
05 / Where are you?
From one-off ChatGPT questions to orders that go into the ERP by themselves.
In operations, the value is almost always in phases 3 and 4: workflows and agents that work in your ERP, TMS or WMS. But you don't start with an agent. You start with a workflow that does one thing well, such as reading in orders. Click through the phases.
Where are you? Click a phase.
- What it looks like
- People use ChatGPT or Copilot for an email or a summary. The handy ones save time, the rest watch.
- What you need
- Licences, a few ground rules and basic prompting skills.
- What it delivers
- Personal and hard to measure. The organization barely notices.
Our advice if you're here
Get your team to the same level in one afternoon. Then you'll see where work piles up.
06 / Knowledge retention
What if the planner isn't there?
Every operational department has one. The planner who knows which customer always wants it a day earlier, which driver doesn't drive to Germany and which machine plays up on Fridays. The work preparation engineer who sees at a glance what's missing from a drawing. That knowledge isn't written down anywhere. Until they're not around.
AI helps capture that knowledge, without anyone having to write a two-hundred-page manual. We take it from what's already there: earlier schedules, old estimates, emails with explanations, notes in the ERP. And we simply ask the expert, in conversations that we record and work out into rules. If the planner adjusts a proposal, we record why, so the next proposal is better.
The result isn't a robot that takes over the planning. It's a proposal that already knows most of the rules of thumb, and a colleague who can take over the planning when needed. The planner can finally go on holiday. And when they come back, there's no chaos waiting.
01Sources
Earlier schedules, estimates, emails and notes in the ERP.
02The expert
Conversations, recorded and worked out into rules.
03Every adjustment
Recorded, so the next proposal is better.
04Result
A proposal anyone can assess, not just the planner.
07 / Work it out
What does AI deliver in your operations?
Three numbers: how many people work on orders, planning and work preparation, how many hours of repetitive work per week, and what an hour costs. Then you see the value per year and the capacity that's freed up.
Calculator
Retyping, searching, answering emails, reporting
Share AI takes over
We always advise on the expected scenario, never on the best case.
Value per year
€ 264,960
Hours back per year
5,888
3.5FTE of extra capacity, without hiring anyone.
This fits package M.
The savings clear the 2x threshold of € 192,000. In the Quick Scan we check it against your own processes.
An estimate based on your input and the assumptions above. Not a quote and not a promise: the real numbers come out of the Quick Scan.
08 / Yourself, tool or partner?
Excel and ChatGPT, an ERP module or an AI partner?
Three ways to make operations smarter. They don't rule each other out. An honest side-by-side:
| Excel and ChatGPT | A module from your ERP vendor | AI Partnership with NinA | |
|---|---|---|---|
| What it does | Ad hoc help with an email, formula or summary | Features within the ERP, such as planning or messaging | Orders, planning, work preparation and reporting across all your systems |
| Reads customers' email, PDFs and portals | · | Often only structured messages | ✓ |
| Works across ERP, TMS, WMS and email | · | Within its own system | ✓ |
| Knows your rules of thumb and exceptions | Only what you paste in | What you can configure | ✓ |
| Who builds and maintains | Everyone for themselves | The vendor or an implementation partner | A dedicated NinA team |
| Fits | One-off questions and analyses | Tidy order flows, one system | Messy input, several systems, knowledge in people's heads |
Swipe sideways to see all columns →
If an ERP module does the job well, by all means use it. We build where the input is messy or the work runs between systems.
09 / How to approach it
From first conversation to orders that come in by themselves.
We build within the AI Partnership. You start with the € 6,000 AI Quick Scan: one day on site, and within a week you know what AI delivers in your operations.
- 01
Intro call
Fifteen minutes
Where does the work come in, which systems does it need to land in, and who is indispensable?
- 02
AI Quick Scan
One day plus a week
We shadow the order desk, planning and work preparation. Three numbers per step, a business case per opportunity, and a twelve-month roadmap.
- 03
First process live
Quarter 1
Often orders or requests into the ERP. Tested on hundreds of real orders from last month, with an employee who approves everything.
- 04
Expand
After that
Planning proposal, work preparation, status queries, reports. Every month you see what it delivers in hours and euros.
10 / Cases
Orders, requests and data with AI, in practice.
11 / Questions
Frequently asked questions about AI in operations.
What can AI do in operations?
In operations, AI can read orders and requests from email, PDF, Excel and portals and put them into the ERP, TMS or WMS as drafts, make a planning proposal, prepare job sheets and estimates from earlier projects, answer status queries and produce reports. The planner and the work preparation engineer decide. At Van Berkel Professionals, request processing alone delivers 192 hours a month.
Can AI put orders into our ERP automatically?
Yes. AI reads the order from the email or attachment, recognizes the customer, items, quantities and delivery date, and creates a draft in the ERP. An employee approves. Orders that deviate, such as an unknown item or an unusual quantity, go separately to a human. What runs without errors for a long time can move to spot checks.
Can AI do the planning?
AI can make a planning proposal based on orders, capacity, agreements with customers and the planner's rules of thumb. The planner moves things around, adjusts and decides. We advise against fully automatic planning: there are always exceptions that only a human can weigh properly. But thanks to the proposal, a colleague can take over the planning when the planner isn't there.
Which ERP, TMS and WMS systems do you connect to?
To ERPs such as Exact, AFAS, SAP and Microsoft Dynamics, and to TMSs and WMSs that have an API or export. Also to email in Microsoft 365 or Google Workspace, Excel and customer portals. We usually build in n8n, with custom work where needed. The condition is access with the right permissions.
What if an order looks different from usual?
That's exactly where AI is good. A fixed rule gets stuck when a customer sends a different format; AI understands what's meant. If the AI is unsure, or something doesn't match the customer agreements, the order goes to a human with the reason attached. That way the exceptions stay with the people who can judge them.
How do you capture the knowledge of experienced colleagues with AI?
By starting with what's already there: earlier schedules, estimates, emails and notes in the ERP. We also hold conversations with the expert, which we record and work out into rules and examples. Every adjustment the expert then makes to a proposal, we record. That way the knowledge keeps growing and the work no longer depends on one person.
What does AI for operations cost at NinA AI?
You start with the € 6,000 AI Quick Scan. Building happens within the AI Partnership: package S costs € 4,500 per month (30 hours), M € 7,500 (50 hours) and L € 12,000 (80 hours), for a minimum of twelve months. What goes live is covered by a maintenance contract from € 245 per month, including AI usage: Bronze runs on your own servers, Silver and Gold on our platform in the EU. All amounts excluding VAT. We only start if the whole partnership delivers at least twice the investment in the first year.
Does AI replace the planner or the work preparation engineer?
No. AI takes over the retyping, searching and calculating in advance, and makes a proposal. The planner and the work preparation engineer assess, decide and resolve the exceptions. What we see: the same people can handle more work, and the department is less vulnerable when someone isn't there.
Further reading
- AI in customer serviceAnswers ready before the ticket opens, including for status queries.
- AI agentsDigital colleagues that plan, carry out and report in your systems.
- AI PartnershipA dedicated AI team that builds, measures and maintains, process by process.
- Automating reports with AIFrom data to report, with the figures from the source.
A department that keeps running when someone isn't there?
Book an intro call. Together we'll look at where orders, planning and work preparation get stuck, and what AI delivers there.
01
Intro call
Fifteen minutes, no strings. Where does work get stuck and what do you want to achieve?
02
AI Quick Scan
One day on site. Within a week, a business case and a roadmap.
03
AI Partnership
A dedicated team that builds, trains and shows what it delivers every month.
160+ organizations · Sessions rated 9.3 · Based in Amsterdam