back to news and insights
Blog

Enhancing operations and efficiency with warehouse digital twins

There are over tens of thousands of goods stored in a typical warehouse, with thousands moving in and out every day. Keeping track of this rapidly changing warehouse operation whilst utilising all the available storage space is extremely challenging and can often lead to errors, inefficiencies and extra costs.

To help warehouse managers gain visibility of the end-to-end journey of goods, Dexory has developed a data driven platform which automatically creates a virtual model of the warehouse using real-time data.

By combining autonomous robotics and advanced perception systems, DexoryView scans the aisles of a warehouse and feeds this information into a digital twin model. Warehouse managers can then analyse this model to track stock, identify discrepancies and optimise the warehouse layout in one easy-to-use platform.

Graphic showing the Dexory robot scanning a rack in a warehouse with the equivalent digital twin model displayed in DexoryView
DexoryView autonomously scans a warehouse and feeds this information into a digital twin model

What is a digital twin?

A digital twin is a virtual representation of a physical object or system. There are three main elements to developing an effective digital twin model:

This ensures that any changes in the physical object are automatically reflected in the digital twin model instantaneously.

‘Once we have generated the digital twin of a warehouse, an autonomous robot navigates the warehouse and scans the locations of racks using lidars and cameras,’ explains Matt Macleod, Head of Software at Dexory.

‘This captures high resolution images and point cloud data which we can then overlay on the digital twin model in real time. Warehouse managers can then investigate any location in the warehouse and analyse the status, barcodes and occupancy of that location – just like they would in the real warehouse, but without all the walking!’

Creating a digital twin of a warehouse

The first stage in generating a digital twin is to complete an initial scan of the entire warehouse. This is achieved with the Dexory robot which is equipped with an array of 3D cameras and lidars. These capture point cloud data which is transformed into a 3D map.

‘Warehouses appear to be the same until you start drilling down and understanding the specific configuration of all the aisles, racks and locations,’ highlights Adrian Negoita, CTO at Dexory. ‘That’s why we use our robot to capture the 3D environment of that particular warehouse so we can then use that information to develop a representative digital twin model for the customer.’

This map is then correlated with the nomenclature of that warehouse, ensuring that location 1 on rack A of the map, matches that of the real warehouse and the Warehouse Management System (WMS). Once the layout of the warehouse has been defined and reconciled with the WMS, the digital twin model is created.

To capture the real-time information of the warehouse, the Dexory robot follows a predefined schedule consisting of several missions depending on which areas of the warehouse the customer wants to scan. The robot autonomously navigates to its first scanning position, automatically extends its tower and begins to scan the racks within an aisle. This data is then sent to DexoryView where it is merged with the digital twin, providing an up-to-date visual representation of the warehouse.

Demo video showing the different views of the digital twin model in DexoryView
DexoryView allows users to understand the stock, status and occupancy of each location within a warehouse


‘The information from the robot itself is useful if you are searching for a specific location, but what is more interesting is comparing the data from the robot with the customer’s WMS,’ says Macleod. ‘This allows DexoryView to analyse what is expected at each location with what is actually there, so any exceptions are quickly flagged to the warehouse manager which they can then correct.’

How does DexoryView help warehouse managers?

The truly autonomous nature of the Dexory robot allows it to scan warehouses at any time. Typically, customers programme the robot to conduct a scan overnight, so that first thing in the morning they have a comprehensive yet clear understanding of the status of their inventory through DexoryView.

The DexoryView software displays all the analytics from each scan for every warehouse site. Users can view a scan summary page which details the total number of locations scanned, any that were obstructed and the percentage of locations that either match the WMS (green), require a manual check (yellow) or are incorrect (red). The software then generates a list of actions in order of priority to help users resolve the most important issues quickly.

To investigate further, users can view the digital twin model of their warehouse in either 2D or 3D. Each location in a rack can be selected and the corresponding barcode and occupancy information is displayed along with the actual photographs and point cloud data from that scan.

Screenshot of DexoryView showing a 2D view of the warehouse with one location selected and the corresponding analytics displayed
The latest barcode, occupancy information, image and scan data is displayed for each location in the warehouse


These analytics help warehouse managers:

‘The fundamental purpose of DexoryView is to compare the latest scanned data from a warehouse with customer data to understand the discrepancies and generate a list of actions that ultimately improves the efficiency of the operations,’ highlights Negoita.

‘Take the typical case of when an item has been misplaced. The next day you will know exactly where it is instead of waiting for an audit which could be in six weeks time,’ continues Negoita. ‘This is just one of the many benefits DexoryView can bring to warehouses. Our customers are now saving up to 40 hours a week of cycle counts and significantly reducing WMS errors, helping them to deliver 100% of their orders to their customers on time.’

EU flag

This company has received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement number 849938