Pharmaceutical LIMS — finding your way to the best solution

July 1, 2007
David Leitham
Pharmaceutical Technology Europe
Volume 19, Issue 7

Vendors who insist on using older and outdated technologies will eventually suffer increased pressure to follow the lead of those who adopt new technologies.

Since they were first developed almost 30 years ago, laboratory information management systems (LIMS) have been used extensively to organize, manage and store analytical data generated within a wide range of laboratories, including pharmaceutical manufacturing, quality assurance and quality control (QA/QC), and in-process testing. Although well-positioned to offer many significant benefits — from sample log-in to reporting, automation to regulatory compliance — LIMS innovation has grown at a rather slow pace. The main cause for this development stagnation has been customer reluctance to adopt newer technologies as they had invested significant effort, time and money in customizing their existing LIMS to meet their specific needs. Additional costs to upgrade and validate newer solutions, particularly for the pharmaceutical industry, led customers to safely stay with enhancements to existing systems.

Yet this model has created frustration. Today, we are seeing customers gravitate to the most complete solution available to reduce the level of customization. Having customized solutions for years, they are looking to vendors to build commonly requested functionality into the core of their systems. Customers are turning to purpose-built solutions to reduce their implementation time, cost and effort. Purpose-built LIMS are specifically designed to include as much as 80% of the required functionality, compared with an estimated 30–40% incorporated in generic or industry-neutral LIMS. Standardizing on industry-specific LIMS makes upgrades far easier than before. New innovations are more quickly accepted and their benefits are more readily enjoyed. Without extensive and expensive customizations, new laboratory data management software can be installed and used at a fraction of the time required for conventional generic LIMS. Validation times are also considerably reduced as standardized products can yield standardized test scripts.

As part of this new era of evaluating LIMS vendors on their ability to develop more innovative software, customers are demanding they keep pace with the emerging need for more standard solutions. In this ever-changing, competitive environment, only vendors capable of adapting and modifying a standardized product offering will be set apart, make profit and eventually survive.

Choosing the right LIMS vendor with the most suitable solutions is certainly not an easy task to undertake. This article highlights the most important evaluation criteria to make an informed choice.

Technological advances

A progressive vendor should have a complete technology plan in place, which will clearly specify their intention to build LIMS based on new technology platforms, such as Microsoft.NET or Sun Microsystems' Java. The primary goal is to demonstrate that the LIMS vendor is innovative and forward-thinking, aiming to offer solutions that use the most up-to-date — yet proven — technology. The plan should take into consideration the resources that are already available to the vendor (legacy, or cherished code), the end goal and a reasonable way to accomplish it.

Key points

Utilizing every new technology is certainly not recommended. Instead, vendors should thoroughly evaluate new technologies before adopting them. Choosing and implementing a new technology can be labour-, time and cost-intensive for both the vendor and the customer. Therefore, the choice of a specific technology should be reasoned, providing a thorough demonstration of the business benefits achieved by using it. In cases where those benefits outweigh the overall cost of the implementation, then the new technology should be chosen.

In general, practically any system can be built based on almost any programming platform. However, each updated version enables increased productivity, while also minimizing the effort spent on changing complex software.

Vendors who insist on using older and outdated technologies will eventually suffer increased pressure to follow the lead of those who adopt new technologies. With IT constantly and rapidly changing, the technology plan must be assigned to a specific person who will be in charge of maintaining and updating the plan through continuous evaluation of new technologies.

Service-oriented architecture

Apart from keeping up with technological advances, it is also equally important for software vendors to predict any future design developments. Solutions developed exclusively on contemporary requirements cannot be effective as they do not encompass enough flexibility to stand up to the varying ways in which the software may eventually change. Such flexibility originates not only from the design of the software, but also from the very processes used to develop it.

You should apply the same rule when choosing a new technology to when you choose a software design. Every new architectural approach must be carefully examined and assessed. There are certainly many proven and efficient approaches that can be implemented, including separation of business logic, user interface and data access code within the system.

Service-oriented architecture (SOA) is one of the most progressive design trends that a vendor can use. This is an architectural paradigm based around packaging large-scale functionality into reusable services. This facilitates assembling applications by marshalling these loosely coupled services, each one of which provides a unit of business logic. This approach is based on open standards and is advocated by both Microsoft and Sun Microsystems as the way to develop enterprise software.

The SOA method is the latest development in distributed computing technologies used for building complex software. The solution has been built improving upon its predecessors including remote procedure call (RPC), common object request broker architecture (CORBA) and distributed common object model (DCOM). SOA places a much greater emphasis on standards and loose coupling between services. Future upgrades of the SOA technology aim to enable vendors to offer services that will be easily integrated with other services within the same organization. As a result, SOA will provide the ability to assemble exactly the functionality needed with standardized services. In turn, this will ensure that vendors are capable of selectively upgrading specific parts of the applications to obtain new features, all with the least possible impact and revalidation on other services.

The SOA method has been adopted by many major software vendors, is governed by the World Wide Web Consortium (W3C) and is advocated by both Sun and Microsoft.

Agile software

The importance of the process employed to develop a new software solution cannot be stressed enough. It is even advocated that the software development process is more important than the technology used to develop the software. As a rule, vendors should always look into new ways of improving their software development processes.

Although time-tested, the traditional 'waterfall' approach is incapable of adapting to evolving business needs. As a consequence, new iterative development techniques have emerged to address this shortcoming. One of the most innovative is called 'agile development' with "agile" referring to maneuverability, but not necessarily to speed, although increased speed can be a usual derivative. Agile development is formed by bringing together different software development practices that, in general, apply time-boxed iterations, adaptive planning and evolutionary delivery.

The main parameter to differentiate agile development from the other iterative techniques is the strong emphasis of the process placed on continuous integration and unit testing. As part of the process, before a unit of code is written, another code is developed to test that unit. While this procedure may be considered labour and time-intensive, it ensures future maneuverability. It allows software vendors to quickly adapt an aspect of their system to changing requirements. Additionally, it helps them to specify the impact this change will have on the rest of the system by implementing the necessary unit tests.

Conclusion

In summary, LIMS users should look for vendors that are capable of justifying why a certain technology should be chosen against another, based on pragmatic analysis. Ideally, vendors should also be in position to either adopt the SOA design method or provide a service-oriented interface to their offerings. Finally, vendors should review and upgrade their software development processes on a continuous basis to keep pace with changing business needs.

It should be noted that functionality is seriously jeopardized when the particular software solution requires extensive customization to meet the customer's needs. This danger has triggered an increasing trend for standardized, purpose-built LIMS. These solutions incorporate most of the required functionality. As a result, customization is minimized thus, reducing risk through easier, quicker implementations, validations and upgrades.

Dave Leitham is director, informatics products, Thermo Fisher Scientific (USA).