Laboratory data is critical to every part of the LNG industry. From soil sample data for upstream exploration, to purity data used downstream to ensure product quality, LNG professionals rely on timely, accurate and well-organised sample data.
Foremost, data must be reliable and defensible. Fortunately, many laboratories adhere to strict protocols to ensure data accuracy, as they cannot afford to have decision-makers question what they are seeing. A reality for many in the laboratory is the requirement that they can prove data quality and do so without sacrificing efficiency. The laboratory’s findings must be defensible, based on proven processes and robust supporting data.
This data-driven approach to laboratory data reliability is not new. ISO 17025, the international standard for testing and calibration laboratory competence, has mandated comprehensive data collection and management since it was first issued 15 years ago. What has changed are the software solutions that laboratories can deploy to meet this challenge.
The importance of data management
A single laboratory test result does not exist in vacuum. Even a commonly executed testing gas chromatography (GC) result is the sum of many complementary parts, from supplier shipment data and staff training records, to carrier gas and reagent quality checks. Ideally, all of these data work together to prevent errors and make results more defensible.
Historically, defending data has been a time and labour intensive process. Laboratory employees had to collect data from dozens of different sources (including the handwritten notes of other staff members) to build a case in support of a disputed result. Even today, many laboratories spend as many as 10 hr/week manually collecting and managing data. This represents a lot of time spent on manual and labour intensive tasks, and sometimes it leads to wasted effort because those laboratory personnel could be utilised on more important activities. Time spent on these manual tasks also represents a significant opportunity cost.
Comprehensive data management software makes it easier for laboratories to defend their data. Laboratory Information Management Systems (LIMS), which were first introduced in the 1980s as basic sample management and data reporting tools, have grown into comprehensive platforms that gather, monitor and manage all laboratory data and records, as well as serving as the integration point for all laboratory instrumentation and, in some organisations, the electronic link to enterprise systems such as Plant Information Management System (PIMS), Manufacturing Execution System (MES) and SAP®. Having all this data well-organised and readily available makes the process of defending laboratory results easier, allowing laboratory employees to redirect time to more productive work, and giving management the assurance that the laboratory data they are relying on for key business metrics are defensible.
GC, a common analytical technique in LNG laboratories, serves as a useful example of the sheer volume of data necessary to support the validity of a single laboratory result. A LIMS allows laboratory employees to easily access this data when and how they need it, ensuring the reliability and defensibility of all test results.
GC is used to determine the composition of a liquid or gaseous sample. It works by vaporising the sample and then sending the resulting gas through a coiled tube, or column, filled with an inert carrier gas, such as helium. The sample’s constituent gases are slowed down by the carrier gas as they pass through the column; each constituent gas is slowed to a different degree, depending on its molecular structure. By measuring the time it takes each constituent molecule to reach a detector at the other end of the column, as well as the volume of each constituent, laboratory professionals can determine the composition of the sample as a whole.
The goal here is not to provide an in-depth explanation of how GC works, but rather to make clear the number of variables that can influence a GC result. Everything from technical errors (e.g. using the wrong column or improperly calibrating an instrument), to staff management errors (e.g. improper training or manual data entry inaccuracies), can render a laboratory result indefensible.
In order to defend their results, laboratories must account for all the variables involved in an analysis and provide evidence that they were executed properly. A LIMS makes this easier by automating the entire data collection and reporting process.
Technical quality data include all variables that are involved in producing an accurate result from an instrument. These data are different to those related to employee performance, which primarily concern the laboratory staff’s competency and adherence to standard operating procedures (SOPs).
For GC analysis in the LNG industry, relevant technical quality data would include reference material certificates, records of approved suppliers, maintenance records, etc. A LIMS provides an entire suite of capabilities for managing these data and for demonstrating the reliability of the final result.
Laboratories require many consumables to operate GC analysers, from syringes to inject samples, to carrier gases that slow the speed of the sample gases as they pass through the column. Rather than retest all of these consumables as they arrive at the facility, most laboratories find it easier to build a network of trusted suppliers.
A LIMS can automate supplier data management by helping track materials as they enter the facility, associating them with their supplier. If a technician needs to defend the results from a test, they can reference the LIMS database to verify that all consumables used were from approved suppliers. If they are, then they can partially rule out consumables as a fail point.
Verifying that a consumable comes from an approved vendor is only part of the equation, as consumables can go out of specification during storage or use. A LIMS can also help solve this problem, as laboratory administrators can configure the LIMS to trigger periodic checks to ensure quality. Together with verification of supplier data, this capability can help laboratories avoid using of out-of-specification consumables in most cases.
Instrument maintenance and calibration
GC analysers must be regularly maintained and calibrated to deliver consistently accurate results. To defend their results, laboratories must also be able to demonstrate adherence to proper maintenance procedures. Laboratory managers can organise and retrieve maintenance and calibration records by instrument, time period and even staff member, directly from the LIMS, allowing them to quickly and easily demonstrate that an instrument in question is adhering to a pre-defined schedule. Verifying this would be more time-consuming and prone to mistakes in a paper-based environment.
A LIMS adds additional value for calibrations that require certified reference materials. In addition to tracking the calibration itself, the LIMS can store reference material certificates that demonstrate the quality of the materials used. This helps mitigate the possibility that faulty reference materials could influence the accuracy of a calibration.
In addition to instrument error, laboratory results can also be invalidated by human error. Human errors in the laboratory originate from issues in one of three categories: training, process and data management. A defensible laboratory result must be able to demonstrate that the laboratory staff that produced it did not inadvertently make errors in any of these areas.
New laboratory staff typically undergo initial training that educates them about the unique processes within their new work environment. However, this training is not static, and regular retraining is necessary as processes change. Training for each employee must be closely monitored to ensure that they are up-to-date, otherwise results from any part of an analytical process in which they participate could be questioned.
Maintaining accurate and up-to-date staff competency records is an onerous process that can be simplified by storing all personnel training records, certifications and permissions within the LIMS. Training is variable and constant, however the LIMS can ensure that it does not jeopardise the laboratory’s quality control processes.
Process quality assurance
Human error is one of the most common causes of failure in any process. Small errors in the execution of a GC analysis can render a result invalid. To prevent this, laboratories must be able to comprehensively monitor the execution of laboratory processes and reference them to individual results.
A LIMS achieves this by storing and automating SOPs to guide staff members through each analytical or operational process. As staff progress through the steps of an SOP, they mark their progress in the system, creating real-time records to document the correct execution of the process steps. Process errors can also be reported to a laboratory manager in real-time, allowing them to correct problems before erroneous results are reported. Multiple SOPs and workflows can be stored in the LIMS, so that changeover for production can be carried out efficiently.
Data entry and transcription
Manual data entry errors made by staff can significantly affect laboratory results. Even if a process is nearly entirely automated, a single manual data entry error can make the result indefensible. A LIMS helps alleviate data entry concerns by automatically collecting and aggregating laboratory instrument data, where it can be accessed by staff or used by other instruments in the laboratory. A vendor-agnostic LIMS is particularly important here because it is capable of interfacing with a broad range of instruments from multiple vendors.
The areas covered in this article by no means comprise the exhaustive list of variables involved in GC analyses. However, they demonstrate that there are many ways for a single experiment to go astray and lead to a disputed result. Defending results is not only about simply verifying inputs and outputs; it is also about documenting complicated processes rigorously at every step, without removing staff from important work.
A LIMS can provide complete visibility into processes and the staff that execute them. Moreover, it can direct processes in a manner consistent with SOP and quality. This complete traceability, from materials supplied to training and instrument calibration, streamlines laboratory operations and provides defensibility. In an LNG industry laboratory, defending results is made easier with the LIMS software.Written by Jeanne Mensingh, Labtopia Solutions (a Thermo Fisher Scientific Partner), USA.
Read the article online at: https://www.lngindustry.com/special-reports/13102015/engineering-defensibility-1453/