NE:ONE automate

NE:ONE automate by Team FRA 2
Group Picture Team FRA 2
Dashboard
Dashboard - Qatar Mode
Change Request View - Qatar Mode
Rule Editor 3
Rule Editor 4
Change Request Approval Screen 2
Rule Editor 2
Rule Editor
Change Request View
Change Request Approval Screen
Change Request Approval Screen 3
List of Rules

The problem your project solves, including which ONE Record challenge(s) are addressed

One of the major strengths of ONE Record is its capability for full data control by the data holder which enables the single source of truth. The data holder, e.g. a Shipper, has full authority over his data and is in charge to decide whether to accept or reject the 'Change Requests' made by the carrier during the goods acceptance. This framework is a significant advantage in digital cargo data management. However, it may also pose with a substantial challenge: the high volume of pending 'Change Requests' can lead to manual processing bottlenecks and potential delays in freight handling.

Challenge 1 - Special cargo validation: Through the automated 'Change Request' process, offered by NE:ONE automate, we enable smooth and fast acceptance processes. Corrections made by the shipper e.g. of the documentation, can automatically be processed by setting up customized rules in the rule engine. Thus, less human interaction is needed and changes can be accepted or rejected immediately.

Challenge 3 - ONE Record AI Challenge: 'Change Requests' are an essential part of ONE Record, that enable the single source of truth approach and allow for true data sovereignty. By creating a One Record 'Change Request' Approval System (ORCAS), we are allowing this mechanism to actually be used in reality. With features like AI correction detection build right in, we are increasing processing speeds and quality even further.

Challenge 6 - Open Challenge: The automation of 'Change Requests' is, as the NE:ONE Server & NE:ONE Play, exactly what the industry needs right now to finally implement the ONE Record standard for data exchange. It enables the industry to leave the old messaging days behind and start this new chapter of a modern and state of the art air cargo product.

Your solution and what it does

With NE:ONE automate, we are creating the first-ever One Record Change Approval System (ORCAS). It empowers data holders to fully manage their 'Change Requests' with advanced automation technology, significantly streamlining the approval process. Effortless control is achieved by setting rules to auto-evaluate incoming 'Change Requests'. By integrating AI modules, such as typo detection, we ensure that manual comparisons, e.g. minor corrections in goods descriptions, are a relic of the past. This is complemented by an intuitive user interface for manually processing any 'Change Requests' that were not automatically handled. With NE:ONE automate, we are addressing a key challenge in ONE Record, which results in enhanced cargo efficiency and an improved customer experience.

The different views and functions of NE:ONE automate:

Main View: NE:ONE automate provides a full list of all the 'Change Requests' with its different status. You can switch this view, to only show “Pending” 'Change Requests'. That way you can quickly see, where a manual review, is still needed. By double clicking a request, you can switch to the details screen
'Change Requests' Details Screen: If a 'Change Request' could not be handled by any rule defined, or the result of a rule is, that a manual check is needed, you can fully view the request on this page. You are presented with a view, that compares the previous and the new version and visually supports you in detecting the differences. Once you have checked the request, you can either Accept or Reject it. When you can’t handle, just be honest and say “Not my Problem” ;)
Rules Engine: This is the heart of NE:ONE automate. Here you can define rules.
- Overview of existing rules. Here you see the current status and further information of existing rules. Create new rules and edit existing ones. Visually set them up with predefined code blocks that represent different logical rules (See the list of our examples rules below)
The Dashboard View: The dashboard is supposed to give you metrics, that show you e.g. show how many 'Change Requests' you received, how many could be automatically handled, how many were rejected etc.

List of good examples for rules, and how they relate to a business process:

When change of City Code into matching Airport Code will be accepted
When update/correction (through an AI) of Goods Description is due to a typo, the 'Change Request' will be accepted
Update of weight within a given tolerance results in acceptance
Update of Product Code results in a manual check.
Carrier updates Security status of an unsecured shipment during Goods Acceptance by screening method XRAY results in acceptance

How did you build it?

Our project was built using a mix of front-end and back-end technologies. We employed Shadcn, ReactJS, Next.js, and Tailwind for a responsive and dynamic user interface. The back-end was powered by Python, FastAPI, MongoDB database with ONE Record API for data management. Blockly added interactive elements, Recharts enabled data visualization, the Microsoft spell check AI for rule enhancement and ChatGPT helped us out where ever we had some knowledge gaps ;). This combination of technologies ensured a robust and efficient system.

What are you proud of?

Having tackled this problem with this first MVP has shown, that an application like this is really useful and can actually help the overall adaption speed of ONE Record.
We managed to implement a visually appealing and actually user friendly product in such a short time
Working as one unit in a cross company team with team members from very different backgrounds with no sleep, but a lot of coffee, caffein, chocolate, food and fun

What is the next step for your solution and how will you take that step?

We see great potential in the development of new rules especially AI-based rules that can make NE:ONE automate an indispensable part of ONE Record.

Some features we did not get to and/or started thinking of during coding are:

Analyzing the efficiency of existing rules in order to derive new rules that increasingly automate the process
Implementing even more predefined code blocks that can be used to create custom rules
Support for all LogisticsObject types
Implementing user management and task delegation
Personalized view for different users of the company