PreFlow: Because 3am Is Too Late to Start Sorting

AI Quick Summary

Today, parcels arrive at Planzer's railway centres as an undifferentiated mass. Every item must be individually sorted under extreme time pressure between 3am and 7am. The sorting belt is the bottleneck, and every second counts. PreFlow proposes moving the sorting upstream through a containerised sorting system deployed either at the sender's location or integrated directly into the Sender2Hub delivery truck. As parcels are handed over, the system would organise them by destination zone. By the time material reaches the railway centre, it would already be pre-grouped and route-ready. The sorting belt would handle exceptions, not the entire volume. This concept needs to be tested. The pilot phase following this challenge would validate the system with a small number of high volume senders, measuring the real impact on railway centre processing time and cost per parcel. If validated, PreFlow may reduce pressure on the night shift and simplify the path to robotic automation. The Sorting Belt Shouldn't Be Doing All the Work Every working night, Planzer's railway centres process roughly 100,000 parcels. They arrive by train from senders across Switzerland, and between 3am and 7am, every single one must pass through the sorting belt, be identified, routed, and loaded into the correct delivery van. The target is over 90 seconds per parcel. The vans must leave by 7am. The margin for error is close to zero. But here is a question worth asking: why are all these parcels arriving unsorted in the first place? Today, when a sender hands parcels to Planzer, they are collected in the order they were packed, loaded onto the Sender2Hub shuttle, and delivered to the railway centre as a mixed batch. A roll container might hold parcels destined for Zurich, Basel, Bern, and Lugano all jumbled together. The railway centre then takes on the entire burden of separating them. This means the sorting belt is not just sorting. It is compensating for a lack of organisation upstream. Every parcel that could have been pre-grouped before reaching the railway centre is instead consuming time, labour, and belt capacity at the most constrained point in the entire chain: the 3am to 7am window. The PreFlow Concept PreFlow proposes a different approach. Rather than accepting unsorted material and dealing with the consequences at 3am, PreFlow would move the point of sorting upstream, before parcels are even loaded onto the train. The idea centres on a containerised sorting system that could be deployed in one of two ways. The first option is a sorting unit placed at the sender's location. High volume senders, such as e-commerce fulfilment centres, process hundreds or thousands of parcels per day. A compact containerised system at their staging area would allow parcels to be grouped by destination zone as they are prepared for pickup. The sender's existing workflow would not need to change dramatically. Parcels would simply be placed into the system rather than onto a single pallet, and the system would organise them into zone-based compartments. The second option is a sorting system integrated directly into the Sender2Hub delivery truck. As the driver collects parcels at each pickup point, the system inside the truck would assign each parcel to the correct zone compartment based on its destination. By the time the truck reaches the railway centre, the load is already pre-organised. The journey from sender to hub, currently dead time from a sorting perspective, becomes productive sorting time. In both cases, the outcome is the same. Material arriving at the railway centre would already be grouped by destination zone. Instead of feeding every parcel individually through the sorting belt, pre-grouped containers could flow more directly toward the correct train wagon or, at the destination hub, toward the correct delivery van. The sorting belt would still exist and still handle exceptions, mislabelled parcels, late additions, low volume destinations, but it would no longer carry the full weight of sorting the entire national volume. Why This Could Matter The potential impact on the railway centre is significant, though it remains to be validated. At over 90 seconds per parcel on the sorting belt, every container that arrives pre-sorted represents dozens of parcels that could skip or accelerate through the sorting process. If a container arrives already grouped for one destination zone, the time saving at the railway centre could be substantial. Scale that across multiple high volume senders and the cumulative effect on the 3am to 7am window could be meaningful. This has implications for cost per parcel, which is Planzer's primary operational metric. Labour on the night shift is the largest cost driver, and anything that reduces the volume flowing through the sorting belt directly reduces the labour requirement. It also has implications for resilience. When a train runs late, the railway centre has less time to sort and load. If a significant portion of the volume arrives pre-sorted, the remaining sorting work can be completed faster, giving the operation more buffer against delays. For employees, less volume through the sorting belt could mean a less intense shift. Whether this translates into meaningfully better working conditions and longer employee tenure is something only a real pilot can determine. What We Do Not Know Yet This concept is untested, and there are real questions that need answers before it can be considered viable. Can a containerised sorting system be made compact and robust enough to operate reliably at a sender's warehouse or inside a delivery truck? What is the right number of destination zones to balance sorting simplicity with operational value? How accurate would the pre-sorting need to be to deliver meaningful time savings at the railway centre, and what error rate is acceptable? What does the system cost to build, deploy, and maintain, and does the cost per parcel reduction justify the investment? Would senders accept having a sorting system at their location, and how does it fit into their existing warehouse layout and workflow? How does the truck-based option handle multi-stop pickups where parcels from different senders are mixed in a single journey? These are not objections. They are design questions that a structured pilot can answer. What the Pilot Would Test The challenge funding would be used to explore the feasibility of this concept through a focused pilot. The first step would be to design and prototype the containerised sorting unit, determining its physical form, capacity, zone configuration, and how parcels are assigned to zones. This could be as simple as a partitioned roll container with labelled compartments or as sophisticated as a unit with integrated scanning and automated routing. The second step would be to deploy the prototype with a small number of high volume senders, testing both the sender-side and truck-side deployment models. The goal is to understand which approach is more practical, more accurate, and more acceptable to senders. The third step would be to measure the impact at the railway centre. How much sorting time is actually saved? How does pre-sorted material flow through the existing infrastructure? What happens when pre-sorting accuracy is imperfect? The pilot is designed to answer the fundamental question: does moving the sorting upstream create enough value at the railway centre to justify the investment and operational change? How PreFlow Connects to the Bigger Picture If the concept proves viable, PreFlow has strategic value beyond the immediate efficiency gain. As Planzer moves toward robotic sorting and automated handling at the railway centre, pre-sorted material is inherently easier for machines to process. A robot handling a container where every parcel goes to the same region can work faster and with fewer errors than one processing a random mix of sizes, weights, and destinations. PreFlow could simplify the requirements for robotic automation, making it feasible sooner and at lower cost. PreFlow also generates data. Every pre-sorted container that passes through the system creates structured information about sender volumes, destination distributions, and timing patterns. This data could feed predictive planning systems, improving resource allocation across the network. The concept is simple. The sorting belt at 3am should not be the first time anyone thinks about where a parcel needs to go. If that thinking can happen earlier, at the sender, in the truck, during the journey, then the railway centre can focus on what it does best: moving parcels onto trains and into vans, fast. Whether this idea works in practice is what this challenge is designed to find out.

Project Idea Metadata

• Project Idea Name: PreFlow: Because 3am Is Too Late to Start Sorting
• Date: 4/5/2026 7:07:09 AM
• Administrators:
  • Elie Maalouli

Project Idea Description

The Sorting Belt Shouldn't Be Doing All the Work

Every working night, Planzer's railway centres process roughly 100,000 parcels. They arrive by train from senders across Switzerland, and between 3am and 7am, every single one must pass through the sorting belt, be identified, routed, and loaded into the correct delivery van. The target is over 90 seconds per parcel. The vans must leave by 7am. The margin for error is close to zero.

But here is a question worth asking: why are all these parcels arriving unsorted in the first place?

Today, when a sender hands parcels to Planzer, they are collected in the order they were packed, loaded onto the Sender2Hub shuttle, and delivered to the railway centre as a mixed batch. A roll container might hold parcels destined for Zurich, Basel, Bern, and Lugano all jumbled together. The railway centre then takes on the entire burden of separating them.

This means the sorting belt is not just sorting. It is compensating for a lack of organisation upstream. Every parcel that could have been pre-grouped before reaching the railway centre is instead consuming time, labour, and belt capacity at the most constrained point in the entire chain: the 3am to 7am window.

The PreFlow Concept

PreFlow proposes a different approach. Rather than accepting unsorted material and dealing with the consequences at 3am, PreFlow would move the point of sorting upstream, before parcels are even loaded onto the train.

The idea centres on a containerised sorting system that could be deployed in one of two ways.

The first option is a sorting unit placed at the sender's location. High volume senders, such as e-commerce fulfilment centres, process hundreds or thousands of parcels per day. A compact containerised system at their staging area would allow parcels to be grouped by destination zone as they are prepared for pickup. The sender's existing workflow would not need to change dramatically. Parcels would simply be placed into the system rather than onto a single pallet, and the system would organise them into zone-based compartments.

The second option is a sorting system integrated directly into the Sender2Hub delivery truck. As the driver collects parcels at each pickup point, the system inside the truck would assign each parcel to the correct zone compartment based on its destination. By the time the truck reaches the railway centre, the load is already pre-organised. The journey from sender to hub, currently dead time from a sorting perspective, becomes productive sorting time.

In both cases, the outcome is the same. Material arriving at the railway centre would already be grouped by destination zone. Instead of feeding every parcel individually through the sorting belt, pre-grouped containers could flow more directly toward the correct train wagon or, at the destination hub, toward the correct delivery van. The sorting belt would still exist and still handle exceptions, mislabelled parcels, late additions, low volume destinations, but it would no longer carry the full weight of sorting the entire national volume.

Why This Could Matter

The potential impact on the railway centre is significant, though it remains to be validated.

At over 90 seconds per parcel on the sorting belt, every container that arrives pre-sorted represents dozens of parcels that could skip or accelerate through the sorting process. If a container arrives already grouped for one destination zone, the time saving at the railway centre could be substantial. Scale that across multiple high volume senders and the cumulative effect on the 3am to 7am window could be meaningful.

This has implications for cost per parcel, which is Planzer's primary operational metric. Labour on the night shift is the largest cost driver, and anything that reduces the volume flowing through the sorting belt directly reduces the labour requirement. It also has implications for resilience. When a train runs late, the railway centre has less time to sort and load. If a significant portion of the volume arrives pre-sorted, the remaining sorting work can be completed faster, giving the operation more buffer against delays.

For employees, less volume through the sorting belt could mean a less intense shift. Whether this translates into meaningfully better working conditions and longer employee tenure is something only a real pilot can determine.

What We Do Not Know Yet

This concept is untested, and there are real questions that need answers before it can be considered viable.

Can a containerised sorting system be made compact and robust enough to operate reliably at a sender's warehouse or inside a delivery truck? What is the right number of destination zones to balance sorting simplicity with operational value? How accurate would the pre-sorting need to be to deliver meaningful time savings at the railway centre, and what error rate is acceptable? What does the system cost to build, deploy, and maintain, and does the cost per parcel reduction justify the investment? Would senders accept having a sorting system at their location, and how does it fit into their existing warehouse layout and workflow? How does the truck-based option handle multi-stop pickups where parcels from different senders are mixed in a single journey?

These are not objections. They are design questions that a structured pilot can answer.

What the Pilot Would Test

The challenge funding would be used to explore the feasibility of this concept through a focused pilot.

The first step would be to design and prototype the containerised sorting unit, determining its physical form, capacity, zone configuration, and how parcels are assigned to zones. This could be as simple as a partitioned roll container with labelled compartments or as sophisticated as a unit with integrated scanning and automated routing.

The second step would be to deploy the prototype with a small number of high volume senders, testing both the sender-side and truck-side deployment models. The goal is to understand which approach is more practical, more accurate, and more acceptable to senders.

The third step would be to measure the impact at the railway centre. How much sorting time is actually saved? How does pre-sorted material flow through the existing infrastructure? What happens when pre-sorting accuracy is imperfect?

The pilot is designed to answer the fundamental question: does moving the sorting upstream create enough value at the railway centre to justify the investment and operational change?

How PreFlow Connects to the Bigger Picture

If the concept proves viable, PreFlow has strategic value beyond the immediate efficiency gain.

As Planzer moves toward robotic sorting and automated handling at the railway centre, pre-sorted material is inherently easier for machines to process. A robot handling a container where every parcel goes to the same region can work faster and with fewer errors than one processing a random mix of sizes, weights, and destinations. PreFlow could simplify the requirements for robotic automation, making it feasible sooner and at lower cost.

PreFlow also generates data. Every pre-sorted container that passes through the system creates structured information about sender volumes, destination distributions, and timing patterns. This data could feed predictive planning systems, improving resource allocation across the network.

The concept is simple. The sorting belt at 3am should not be the first time anyone thinks about where a parcel needs to go. If that thinking can happen earlier, at the sender, in the truck, during the journey, then the railway centre can focus on what it does best: moving parcels onto trains and into vans, fast.

Whether this idea works in practice is what this challenge is designed to find out.

Today, parcels arrive at Planzer's railway centres as an undifferentiated mass. Every item must be individually sorted under extreme time pressure between 3am and 7am. The sorting belt is the bottleneck, and every second counts.

PreFlow proposes moving the sorting upstream through a containerised sorting system deployed either at the sender's location or integrated directly into the Sender2Hub delivery truck. As parcels are handed over, the system would organise them by destination zone. By the time material reaches the railway centre, it would already be pre-grouped and route-ready. The sorting belt would handle exceptions, not the entire volume.

This concept needs to be tested. The pilot phase following this challenge would validate the system with a small number of high volume senders, measuring the real impact on railway centre processing time and cost per parcel. If validated, PreFlow may reduce pressure on the night shift and simplify the path to robotic automation.