Intralogistics automation: what a sequencing process delivered in six weeks teaches
Intralogistics automation pays off particularly where manual, recurring transport and sequencing steps set the pace. A documented case from VW's Gläserne Manufaktur shows, according to the company, that such a process can be fully automated with transport robots, a cobot and adapted load carriers, implemented in six weeks.
For decision makers in brief
- Context: The case is an external industry example from our analysis of 52 published practitioner conversations (September 2025 to August 2026). It is a company statement, not a project we were involved in.
- Effort: The bottleneck was not the technology but clarifying the physical workflow. Load carriers, routes and handover points had to fit the automated process.
- Cost: According to the company, CO2 and cost per sequence fell by 50 %. For your own planning, the short implementation time matters most, because it lets you calculate the business case in months rather than years.
- Risk: If AI in robots takes over a safety function, such as collision avoidance, stricter requirements apply. This needs to be clarified before the project starts.
- Transferable: The pattern, not the solution. First clarify the process, then automate, then show a measurable result quickly.
The case: a sequencing process in the Gläserne Manufaktur
At VW's Gläserne Manufaktur, a previously manual sequencing process was automated. Autonomous transport robots, a cobot and wooden load carriers were used. The following details come from a published practitioner conversation and are labelled as a company statement in our industry report "AI in Manufacturing".
| Feature | Detail |
|---|---|
| Company | VW Gläserne Manufaktur |
| Starting point | Manual sequencing process |
| Means used | Autonomous transport robots, cobot, wooden load carriers |
| Degree of automation | 100 % |
| CO2 per sequence | reduced by 50 % |
| Cost per sequence | reduced by 50 % |
| Implementation time | 6 weeks |
| Field of application | Logistics, automation |
| Type of source | Company statement, own analysis |
Two limitations belong to an honest assessment. First, the information is based on the published summary of the conversation, not on an independent review. Second, the source gives no details on investment volume or quantities. The case therefore serves as evidence of feasibility and speed, not as a template for a cost calculation.
Pattern 1: speed is the real currency
Only 14 of the 52 conversations analysed mention a result with a figure at all. Remarkably often, this figure does not prove the technology but the speed of implementation: six weeks at VW, ten months to start of production in a platform project, under 24 months per plant at Progroup.
This is no coincidence. If you are championing an initiative internally, the time to the first result convinces more than percentages. Economics add to this: according to BCG 2026 (consulting study), payback periods for AI projects in high-wage regions such as DACH are materially shorter than in low-wage regions, where comparable projects take two to eight times as long. The high cost pressure in industry is therefore also a strong argument for automation.
Predictable speed depends on a clear framework. The page on our AI project fixed price describes what such a framework with 72 hours to the first pilot and 4 to 10 weeks of implementation looks like.
Pattern 2: the lever lies before the AI
Several of the strongest figures from the conversations do not come from AI projects but from process work. The VW case, too, is primarily an automation project built on a reorganised workflow. Two further documented cases from intralogistics and material flow control show the same pattern:
| Company | Measure | Result | Type of source |
|---|---|---|---|
| Magna Exteriors | Automated picking in automotive logistics | 97 % fewer walking distances | Company statement, own analysis |
| RUWAC Industriesauger | Switching five halls from push to pull | Lead time and work in progress each reduced by 50 % | Company statement, own analysis |
| VW Gläserne Manufaktur | Transport robots, cobot, wooden load carriers | 100 % automation in 6 weeks | Company statement, own analysis |
Several interviewees state the same rule: first clarify the physical process, then automate. The industry report includes it as a separate step in its roadmap. For intralogistics this means that routes, handover points, load carriers and sequences are defined before any software decision. A robot that maps a disorganised workflow only automates the disorder.
Intralogistics automation: where AI comes in
Not every automation is AI. Transport robots and cobots often work with classic controls. AI comes in where planning, sequence and demand are continuously recalculated. According to Deloitte AI in Manufacturing 2026 (consulting study), 49 % of the manufacturers surveyed use AI in logistics and the supply chain, after quality at 62 % and production at 57 %.
In the industry report, this area belongs to field of application C, planning and supply chain. The prerequisites are clearly named there:
- Reliable master data in the ERP, because no other field is as unforgiving of poor master data.
- An order and sales history over several years, including outliers.
- Realistic replenishment times per item, not the wishful values maintained in the system.
- A single target metric, such as service level, stock coverage or delivery reliability.
A typical entry point is a defined item group, initially in parallel operation. The measure of success is not model quality but the measurable effect at the same service level. Our article on AI use cases in manufacturing shows which other fields exist and how mature they are.
Safety: when robots become relevant for regulation
As soon as AI takes over a safety function in a machine, the situation changes. The industry report gives examples such as camera-based monitoring of protected zones, AI-supported collision avoidance for cobots and adaptive speed limitation. In our reading, such machines fall under Annex I of the Machinery Regulation (EU) 2023/1230, which applies from 20 January 2027. Self-certification then no longer applies, and a notified body is required. From 2 August 2028, the high-risk obligations of the AI Act under Annex I are added.
This assessment is not legal advice. The specific classification depends on the function in the product and should be reviewed by a lawyer before the conformity assessment. Important for planning: in this reading, transport solutions that do not intervene in the safety chain generally remain outside the high-risk classification.
What you can take from the case
The VW case cannot be copied, its pattern can. For your own intralogistics automation project, five checkpoints emerge:
- A manual, recurring process with a clear metric, such as walking distances, lead time or cost per operation.
- A baseline taken from the system before the start.
- A clarified physical workflow before robotics or software are procured.
- An assessment of whether a safety function is affected.
- Defined operations, so that systems are maintained, monitored and adjusted after go-live.
Many projects underestimate the last point. Regulated AI operations as a managed service define who monitors and retrains models when the product range, cycle times or layout change. Our page on AI for manufacturing companies describes which requirements apply specifically in production environments.
