Problem: Prototype did not have software that was suitable for production.
Our customer CryoXtract was in the process of turning university technology for coring frozen biological samples for biobanking extraction applications into a commercial product. Customers tasked with extracting DNA from biological material, especially material that has been stored in a freezer over many months are challenged with obtaining precise volumes of biological material from small tubes and require the ability to reprocess the samples for future testing. The technical challenges in allowing this type of processing require advanced vision analysis to detect material within the tube and determine which portions have already been removed. Perhaps the biggest challenge CryoXtract resolved was creating a system that would ensure that the samples provided maintained extreme cold temperatures throughout the process because simply thawing material with DNA and then refreezing this same material would destroy much of the remaining DNA available for future sampling. The technology had been proven to work and a prototype had been built for trial customers throughout the world.
When we were approached by their team the product was being produced at a manufacturer and getting ready to ship. However, the current state of the system was developed using control software which had many technical problems. Fundamentally what was a lab experiment turned prototype had evolved with no clear direction on how to architect the machine application and allow for future expansion based on customer feedback. At its core there were a few major issues that had to be addressed prior to shipping the first unit:
- The load time of the software took minutes and it was not developed to boot up upon system power up.
- Machine errors proved fatal when they occurred and the user would need to reboot the system. The user interface lacked many critical features for user feedback and there were no diagnostic utilities for servicing the machine in the field.
- A more intuitive user interface was required to better present information to all user types. Machine failures needed to be handled in a way that a novice user could recover from errors generated. Also diagnostic utilities would need to be developed to better troubleshoot system errors both by the customer and field service professionals.
- Each customer application required variances in the coring process
- Customers stored samples in different tubes which meant the diameter and depth of the material contained may vary requiring abilities for users to define various tube dimensions.
- Customers required precise volumes of the material extracted which meant the vision software and mathematics of choosing a coring pattern to retrieve the correct amount of material would need to be controlled
- Some customers required modifications to the process as an example some tubes needed to capped and decapped while some customers provided tubes with caps already removed. Basically based on the material used and its containing element amongst other constraints no single control process would work for all customers and the current software lacked all of these features
As a result Origo Automation was tasked to evaluate the existing software architecture and propose a solution that resolved the current challenges and allowed CryoXtract to provide a reliable beta release to their new customer base. The software had to autorecover if possible from machine errors. It required detailed event logs for trouble shooting issues in the field. And the user interface required an overhaul to allow certain initial features while providing flexibility for future expansion. Our analysis revealed that a complete rearchitecure was needed to solve the challenges and allow for easy future upgrades. We proposed a new architecture for their software application and provided CryoXtract with a new design that would accommodate the requirements. We identified the critical must have features for this first release of their product in addition to the features that were not necessary but would eventually require integration. We then provided CryoXtract with a detailed schedule that would allow us to release a working prototype with required features that would meet their deadline and also provide a path for enhancing the software as their users provided feedback and new requirements.
Ultimately we were able to convince CryoXtract that we could provide this new software architecture by their deadline and we were awarded this difficult task of turning a faulty prototype into a reliable product that had many new feaures . Within3 months we developed a software architecture that spanned 3 processing cores, provided a scripting feature for their engineers to modify the control process as needed for their customers, and enhanced their user interface capabilities.
After the release of the first few units customers were providing feedback and feature requests which Origo Automation ultimately developed and maintained for CryoXtract. We were able to provide reliability of operation with minimal system down time. We enhanced the situation error handling to provided auto recovery from system errors as they occurred. Non-recoverable machine became easy for field servce to troubleshoot from the deteaile machine event logs. Ultimately Origo Automation was able to provide the first working prototype on time and continued to work with CryoXtact as a partner to develop new enhanced features and provide support to both customers and field service providers for many years.