Better data management: moving beyond LIMS

Ensuring that you can make the most of the reams of data generated in a modern laboratory you need a system to integrate data from a myriad of sources

ONE of the key challenges faced by organisations today is their inability to turn the vast amount of laboratory data that is generated into useful information that enables them to make timely and effective decisions. With multiple applications across the enterprise generating reams of data that all sit in separate silos, aggregating and mining this data is a very real and complex problem. This complexity is compounded even further when a laboratory manager or director has to manage multiple teams or groups of work. All of these are barriers to making effective business decisions and increasing ROI.

A common reason for purchasing a Laboratory Information Management System (LIMS) is its ability to integrate with laboratory instrumentation and enterprise systems for better data management. And yet, according to a worldwide market research study on LIMS users conducted by SDi (Strategic Directions International, Los Angeles, CA), approximately 70% of the respondents manually enter instrument data into the LIMS. A coherent strategy that can integrate data from LIMS, CDS, and other sources (e.g. Enterprise Resource Planning or ERP, and Manufacturing Execution Systems or MES) across the enterprise is a key business driver today. 

"Making information available as soon as possible within a manufacturing operation plays a major role in making any production process efficient, particularly when that information indicates a change within the process performance"

In many industrial laboratories, chromatography provides a significant amount of critical data, and so it is vitally important that all this data is processed and made fully available as soon as possible in the LIMS. But, since instrumentation comes bundled with specialised workstation software, and laboratories typically use more than one type of instrumentation, supporting multiple software products can lead to an increased burden on IT. Chromatography Data Systems (CDS) were conceived as a standardised enterprise-wide solution to simplify this complexity and support multiple and diverse instruments. These systems have increased productivity and reduced costs associated with continuously implementing and validating multiple instrument-specific or laboratory-based software solutions. This is particularly true in laboratories that have to conform to stringent regulatory requirements, such as pharmaceutical companies.

Full integration of LIMS and CDS is the next step in laboratory data integration and management. This reduces the complexity and expense of implementation, validation, maintenance and training when compared to managing multiple software systems. Productivity gains are also achieved through the consistent reporting and sharing of data. A well-integrated LIMS and CDS is designed with the chromatographer’s workflow in mind, allowing multiple levels of user interface, intuitive and user-friendly operations, and automating the flow of LIMS-generated work requests into the CDS and CDS results back into the LIMS.

In a manufacturing environment, the creation of a production order in an ERP system is a key step in any large-scale production process. This action cascades through the manufacturing operations and triggers activities within domain-specific systems such as LIMS and MES (Manufacturing Execution System), in addition to updating product specifications or determining production quantities. As soon as the ERP triggers a change in manufacturing requirements, the LIMS is able to update sampling and testing schedules and specifications accordingly and the management of laboratory resources is immediately available for the new scheme of work. Also, data does not need to be replicated in multiple systems, since this is the whole advantage of integrating a purpose-built LIMS in the manufacturing environment.

Making information available as soon as possible within a manufacturing operation plays a major role in making any production process efficient, particularly when that information indicates a change within the process performance. Timely and accessible data are vital to preventing product loss or the production of out-of-specification materials and reducing the likelihood of lost revenue.

In addition, on-line analysers in the plant are employed to make key decisions as quickly as possible, and laboratory results are often used to verify data, especially if a parameter is out of specification.  If a significant amount of time elapses before laboratory results become available to the plant operators, these results may be irrelevant or not representative of how the process is currently performing. Therefore, the rapid delivery of laboratory results is key to ensuring that the data is relevant. Having a direct interface between the LIMS and MES allows the production facility to track the status of its on-line results and make timely decisions.

Making laboratory results available, however, is not the only requirement of an integrated system, since it is often the case that the laboratory and the plant operators have different frames of reference and often refer to data using different terminology. Passing information from one system to another needs to be done in a way that considers all the original source parameters, in their sometimes different forms. Since manual data calculations can lead to delay and potential errors, shared information should also be formatted for its recipients as a function of the interface.

Regardless of the industry, the need to integrate information across the enterprise is clear. How this integration is achieved is the current challenge for many organisations and software vendors. One way is by adopting a service-oriented architecture (SOA) and taking advantage of the integration capabilities it can provide.

One of the benefits of an SOA is its ability to provide application-specific results and information to other applications on demand, much like an external application programming interface (API) does. The difference is that, today, SOAs are usually based on Web services, which use universal, well-known protocols for communication. By supporting this architecture, applications such as Thermo Scientific LIMS and CDS can access, via web services, applications such as Microsoft BizTalk and SharePoint.

Microsoft BizTalk, which has support for communicating between 400+ external product formats, goes a long way toward facilitating integration of disparate systems such as PeopleSoft Enterprise, SAP, JD Edwards Enterprise One and many more. Thermo Scientific integration technologies provide a bus-like, configurable, distributed communication pipe between any two systems. They support integration of both software systems such as LIMS or CDS, and hardware systems such as laboratory instrumentation.  In addition, the Microsoft SharePoint platform has built-in support for creating “composite” applications, enabling the development of portal-like Websites, which allow data from multiple back-end systems (such as LIMS, CDS, ERP, MES or any other system) to be converted to key business metrics or other critical information. These tools enable easy integration of what are traditionally separate applications into a composite application, and provide the ability to aggregate and surface data that is relevant. From the user’s perspective, he or she is given the flexibility to see what they want to on their desktops, and make accurate an quick business decisions.

In the future, these technologies will further integrate with Microsoft BizTalk, so that any software or hardware that can communicate with Thermo Scientific LIMS and CDS will be able to communicate in turn with any other product that Microsoft BizTalk supports. This will open up the 400+ products that BizTalk supports, expanding customers’ capability to share and communicate. The ease of data mapping and integration will ensure a far easier, less labor-intensive flow of data across the enterprise.

The world of laboratory informatics is changing. As laboratories look to streamline the flow of information, having multiple disparate systems with minimal to no integration is no longer an option.  The shift to a laboratory fully integrated with the enterprise, such as the one described here, will yield a greatly enriched user experience that allows individuals and organizations to not only manage and capture their data more efficiently and securely, but also to simplify their daily workflow. At a higher level, it enables management to make decisions based on current and pertinent information. This ability to turn large quantities of data from disparate sources into valid business knowledge is critical. Companies using these technologies and techniques are the ones which will grow and gain competitive advantage in today’s volatile business environment.

By Trish Meek, Séamus Mac Conaonaigh, and Colin Thurston. Mrs Meek is director of Product Strategy, Life Sciences, Mr Mac Conaonaigh is director, Technology, and Mr Thurston is director of Product Strategy, Process Industries, at Thermo Fisher Scientific.