Product Request: Name-Value Pair Enhancement for Batch Record Entry
Request Summary:
Enhance the Tulip platform to support a more user-friendly method for entering batch record values using a name-value pair approach. This enhancement should allow operators to enter values directly into separate fields on the screen, similar to a paper batch record, without the need to navigate through tables.
Key Requirements:
Name-Value Pair Support:
Implement a methodology where each parameter (name) and its corresponding value can be displayed and edited in separate fields on the screen.
Ensure that these fields are easily accessible and modifiable by the operator without navigating through multiple rows or tables.
User Interface (UI) Improvements:
Design the UI to display parameters and their values in a clear and organized manner, similar to a paper batch record.
Avoid using interactive tables for data entry, as they are cumbersome and not user-friendly.
Scalability and Flexibility:
Ensure the system can handle a large number of parameters and values without performance degradation.
Allow for easy addition of new parameters without requiring significant changes to the underlying data structure.
Integration with Existing Systems:
Ensure compatibility with existing Tulip features and workflows.
Provide a seamless transition for users currently using tables for data entry to the new name-value pair system.
Justification:
User Preference: Operators prefer entering values directly into separate fields on the screen, which is more intuitive and aligns with their workflow.
Improved User Experience: A name-value pair approach reduces the complexity and potential for errors associated with navigating through tables.
Scalability: This approach allows for easier management of a large number of parameters, which is essential for extensive use cases like Electronic Batch Records (EBR).
Scenario: How do I read data back from a completion record. I enter data, I move to next step or additional steps and I want to go back to previous step to review data and potentially change it. How can I do that? Solve was to do this with a completion record, but how do we do that?
Conclusion:
Implementing this enhancement will significantly improve the user experience for operators, align with their preferences, and ensure the Tulip platform can handle complex and extensive data entry requirements efficiently.
Thanks for the suggestion here. I want to make sure the team understands this fully, what are the limitation you see to using input widgets for operators to enter values needed for the batch record? To use a concrete example, if someone need to input a work order value, they can do so with the input widget, and that value is then saved to the variable associated with that input widget (which I think would accomplish the name-value pair scenario).
@andybmac I am going to lean on you here to give a bit more info and maybe share a screenshot or recording to help better define this if you could pretty please!
@Beth@ZAP - I will add more detail, but in 2 weeks time as I am just leaving for vacation!
In the mean time - the issue is that the completion record (and therefore variables) is not designed for reading and updating the extensive number of values that we would have in a large pharma batch record - and the 200 column limit to a single table certainly wouldn’t cope, which is why traditional MES store values in a relational structure/name value pair.
If you look at a typical biotech stainless facility -with the number of upstream (Fermentation) and downstream (Purification) process stages and every step requiring the functionality to go back and display values as entered and allow for a correction then the issue becomes a question of scale. (A typical large scale facility would have multiple cells, each with multiple trains and each of them with multiple bioreactors - circa 15,000 litres etc).
I think this highlights a very key challenge with the system and scalability we face.
I don’t know the future of the completion record and record history widget, but it can be used to highlight what end users are expecting. It needs to be organized for humans to read and interact with the data to allow for handling mistakes. We also need to know what to present in an audit situation. Some of our customers have to share the electronic batch records with customers and partners.
When showing the record history record end users are expecting something like depicted below.
If the end user point to an entry in the record history a “Comment” button would appear. This could have a trigger which would allow us to capture the interaction with the record and post an entry in the audit log with a link to this record.
If the user points to a specific data point it could be highlighted with an “Edit” or correction button which allows for correcting and or commenting on the value.
Dont get hung up in the integrity test example shown for illustration only and impacts on correcting data. I know I am simplifying this a lot and I do appreciate that corrections is a really complicated topic since commenting/editing data easily impact other records.
But having some way of correcting flawed data is needed. If analytics are built on top of flawed data they are wrong too (though you have added a possibility to exclude specific completions).
If the concept of completion records are expected to be the vehicle for controlled/scalable data capture and presentation going forward concepts like this needs to be thought into the solution. If Tulip chooses a different path this doesn’t remove the need for interacting with the data.
