CHANGE MANAGEMENT | LABORATORY INFORMATICS GUIDE 2015
structured way. Raw data represents a set of unstructured data points. A data file without context or meta-data information is meaningless. Adopting automated, self-documenting, data
capturing processes at source increases the value of scientific data. To re-use experimental data in other processes requires accurate meta- and context data. Systematic tagging of metadata to objects will significantly facilitate more effective searching. This is happening already in the consumer
The laboratory delivers scientific evidence for many departments
achieve with our scientific processes. Some of our scientific processes have become a matter of habit and have become deeply embedded in our daily routines. Adopting a client-focused (requester) mind-set
may help us. For example, a patent lawyer is not interested in all the scientific details of how you created new discoveries. He needs a simple proof of evidence, and a confirmation that standard operating procedures were followed. Internally, a detailed document should be available, in case there is a need for additional evidence. When a marketer is exploring competitor’s products, he is unlikely to be interested in the absolute amount of a certain chemical concentration, or a physical attribute. He needs to understand it only relative to his products. In my market observation, I have observed that the language we speak is not always in line with the expectations of our lab requesters.
RIGHT, FIRST TIME As with any language, effective communication relies upon context. Metadata that enables one to understand whether one is comparing apples with apples, or apples with oranges, is critically important. Across all sectors, within every R&D
process, data is generated in ways that carry vital contextual information for all consumers of that data. This context could be as simple as temperature variation, or as complex as a genome, but capturing instrument, sample preparation methods, analysis parameters, or observational data is essential to R&D. Even a ‘simple’ measurement is not meaningful unless the experimental and analytical conditions are specified, and any subjective observations and
conclusions are associated with the data. Overall there are three basic principles to
optimising the integration of data: 1. Capture the data at the point of origin to eliminate human error and to reduce system complexity;
2. Simplify and implement self-documenting processes to eliminate transcription errors and avoid unnecessary retyping of data. In a recent survey, 32 per cent voted that data integration in a paperless laboratory will eliminate manual entries and data transfer; and
3. Ensure that metadata is captured in a Laboratory change process categories
Mind-set and change management
l New scientific data consumers may have different metadata tagging needs l Capture data at the source. Self-documenting processes l Align cross-boundary processes. (What’s-in-it-for-me syndrome)
l Paper will be used as an intermediate medium. Archiving and searching experiments will be paperless
Internal and external process integration
Consolidation of systems
l Integration of CROs’ and CMOs’ processes to leverage collaboration efforts l Electronic regulatory processes requires rethink
l Redefine retention time policies. Investigate purpose (e.g. regulatory storage vs re-use)
l Impact of simplifying and consolidating legacy systems requires cross- departmental rethink. The role of IT to focus on security, data standards (controlled vocabularies), and system availability
l Reduction in overlap of applications will reduce operational costs and validation effort
l Adopt lifecycle management l Think in terms of ‘capabilities needed’ instead of ‘products to buy’
Adoption of new technologies and processes
l New architecture for data storage l Effectiveness of tablets and mobile devices l Change of financial planning. Opex vs Capex l Impact of SaaS vs traditional model l Legal landscape impact of accessing Big Data in scientific communities
www.scientific-computing.com/lig2015 | 5
electronics industry. For example, when capturing a photograph or film, a smartphone will by default systematically add metadata to the object. Examples include GPS location, weather conditions – sometimes even the user’s personal condition such as heartbeat, etc. The scientific community should expect to
see the introduction of similar developments. Modern balances may automatically include temperature and humidity while recording and transmitting the weight. Balance and titration manufacturers are adding value to their instruments by implementing approved and pre-validated methods in their firmware. Chromatography data handling systems (CDS) can add instrument parameters to raw data files, such as temperature of the oven, pressure of the mobile phase, and frequency of data collection. Modern systems also report run-time deviations of instrument settings. It may sound a small step, but all this will significantly reduce variability and validation
➤
Courtesy of The Edge consulting & Industrial Lab Automation
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36