An old problem. A new Solution.

As we move towards a society increasingly influenced by the internet, and in-turn eCommerce, we have witnessed an exponential rise in the number of goods shipped from Asia.

Online, the plethora of options offered by any of the dominant eCommerce platforms gradually guide of down a familiar path.  The method of delivery, variety of choice and crucially, the incredibly low prices inadvertently have led us to China as the point-of-origin for the purchases of a fantastic 39 Million Europeans in 2017. Sounds good. So what’s the problem? Well crucially, this ever-growing deluge of parcels coming to Europe from Asia brings with-it the conundrum of how to effectively, and moreover profitably, handle such an enormous amount of troublesome packages without them literally piling up, at massive expense to the European parcel operators.

Asian Parcels. A mounting problem.

In 2017, Alibaba alone, shipped over 560 million parcels to the EU mainly via Alibaba Express. From a country with a history of notoriously poor labelling, the fact that 81% of these parcels had employed the Cainiao Network in an attempt to standardise labels is good news and viewed as a significant step in the right direction. The truth is, however, that a substantial proportion of these parcels still suffer from a variety of issues when it comes to the automated sorting process. 

Firstly, the packages themselves tend to be erratic in size, shape and weight which can cause multiple problems when trying to present the label in the correct position to a scanner.  The CN22 & CN23 customs labels and barcode system often suffer from poor printing, with crucial information missing or the ink smudged.

Lastly, even when a parcel is a standard shape, flatly presented to the scanner, with precise, clear custom labels; The lack of standardisation in senders’ address label formatting makes a near-impossible read for the more common scanners. Add to this; the fact the remaining 19% of the 560 million parcels have no standardised formatting at all, and it becomes a daunting prospect for any modern European sorting hub.

When all these factors take effect, is an obvious conclusion that ever-increasing exports of goods from Asia are causing a significant handling challenge for European postal operators. And it’s unavoidable. European operators are obligated, under UPU regulations, to process any foreign import stream as effectively as possible.

The lack of optimised automation OCR available on the market with the capability to handle this complicated import stream means that vast resources need to be immobilised to predominantly manually sort these parcels at considerable cost to the postal operators concerned.

With all this mind; The love of the challenge and an opportunity to demonstrate our pioneering edge, Prime Vision have designed an innovative solution that directly addresses virtually all of the persistent issues that have historically prevented this growing, and potentially very profitable, mail stream from being automated. 

An advanced solution.

The eCommerce Parcel Reader is based upon Prime Visions awarding-winning OCR technologies; however, our new innovations have taken this system to the next level.  These developments enable us to read data across multiple labels including all address labels, customs labels and recipient details. Whether a postal operator is already employing Prime Visions systems or not, the versatile nature of the solution allows our eCommerce Parcel Reader to be offered as an additional recognition engine to any established coding and sorting operation.

Rather than taking the obvious solution in attempting to identify and define the numerous types of labels that may (or may not) be on a typical parcel, the Prime Vision solution approaches the problem differently. Our solution focuses on detecting, interpreting and validating only the information necessary for processing and sorting. Looking for the relevant blocks of data, hidden in a sea of confusing and useless clutter.  

Employing the latest machine learning techniques, our development team, have created a solution that can first interpret the type of mail (in this case Asian) and then identify ‘segments’ of relevant data to establish the location of only the necessary text. Once pinpointed, further machine-learning and advanced OCR techniques combined with our parcel OCR libraries ‘clean’ the address line for conventional recognition. 

New functionalities are currently under development.  For example the recognition, scanning and uploading straight into the system of custom labels data and costs, ensuring correct duties are applied, and ongoing parcels processed faster.

Turning a negative into a positive

With the fantastic growth of the Asian parcel market of the last few years, Prime Visions eCommerce Parcel Reader promises to be a powerful addition to any current OCR solution. The solution offers an opportunity to postal operators to convert the incredibly busy, but labour intensive, Asian parcel stream into a profitable revenue stream by sustainably reducing their manual handling cost.