A practical approach to a nalytical lab scheduling

Time and money can both be saved by careful planning and scheduling of work within the analytical laboratory, as Alex Howard and Guy Malchi of Tefen* explain

The trends in the pharmaceutical industry, as with most industries, are towards lower cost, with higher quality and greater added value in order to sustain competitive advantage and hence increase market share. In an industry where quality has become a focal point of activity it is still rare that planning and scheduling of the activities in QC labs is carried out. Despite the complexity and variability of the activities carried out in QC Labs, it is possible to create a database of tests and times that can be used for planning resources in the long and medium term. It is also possible to prioritise the tests to be performed by analysts and match them to their individual skills, thus providing a scheduling tool that details the activities to be carried out on a daily basis.

The benefits of such a system are clear. It provides visibility of the QC resources that are available, the associated costs and the future capacity required for increased production, enables effective management of resources and the ability to schedule effectively on a daily basis, and ∑ enables efficiency improvements (by enabling batching etc.) and helps to reduce cycle time.

The pressures of today's manufacturing environment and the strict quality regulations placed on the pharmaceutical industry mean that the quality control (QC) function is an essential step in the process. QC Labs in industry (and other types of analytical laboratories) are under increasing pressures to operate in the same way as lean production style environments. There are the usual pressures to cut costs, in parallel with demands for shorter cycle times and greater flexibility to deal with increasing customer demands. QC Labs are expected to deal with daily fluctuations in workload and heavy demands on cycle time, all within budget. There is a surprising lack of systems and tools available on the market to help QC Managers with the complex planning and scheduling that is required. This article discusses the importance of planning and scheduling in analytical laboratories, the associated problems and the benefits of introducing such a system. It also describes a methodology and tool that can be used for resource planning and daily scheduling of laboratory activities.

QC Labs and, indeed, most analytical laboratories are environments that typically have any number of skilled analysts testing large batches of samples. This usually involves a large variety of activities for a range of complex tests. The tests are often carried out on HPLC machines or similar, but it is the skills of the analysts that are usually the limiting resource. There are also invariably a large number of tasks extra to the testing of batches of products, which are very difficult to quantify (paperwork, cleaning, administration, development work etc.). Such laboratories exist in bulk pharmaceutical, biotech and even hospital environments.

why planning and scheduling?

Planning and scheduling have always been essential functions of manufacturing operations. The modelling of requirements based on long term plans, forecasts and projections is essential for the planning of resources and materials in manufacturing operations. For QC Labs that are typical of those described above, there are few, if any tools to aid the planning and scheduling of such environments. Often the LIMS system in the laboratory has a module that claims to be able to provide some sort of planning functionality, but it is either not used, or found to be woefully inadequate. The problem is usually the complexity of the environment that is to be modelled in the system.

The problems suffer by QC Labs are invariably similar and usually include:

1. Limits on resources and budget constraints;

2. Wildly fluctuating demand;

3. Pressures on cycle time;

4. Difficulties planning & scheduling.

Limits on Resources and Budget Constraints

Every year, the QC manager is required to justify the expense of analysts, and is required to cut costs wherever possible while dealing with increasing production volumes. This is a very difficult task as there is usually no supporting data on available capacity or future capacity requirements. It is very difficult to justify the extra resources required, and it is often assumed by the rest of the organisation that the QC Labs have infinite capacity.

available QC resources

There is a need for a tool that clearly demonstrates the QC resources that are available and the future capacity required for increased production.

Wildly fluctuating demand

The demand placed on a QC Lab fluctuates wildly on a day-to-day basis. This is typified by the graph in Fig. 1.

The problem of managing such demand is difficult as there is no visibility of what the peaks and troughs will be and no accurate knowledge of where the available resources line is. While the lab is under-loaded, there is no visibility that this is the case, and no way of knowing to what extent. While it is overloaded, it is clear how much extra work there is (as it visibly piles up) but it is still difficult to manage. The QC manager must find the extra resources and analysts are regularly asked at short-notice to do overtime. In such an environment, it is impossible to work efficiently as the high priority batches have to be worked on at all times, rather than effectively batching the tests.

There is clearly a need for a tool to manage resources as far as possible and schedule effectively on a daily basis.

Pressures on cycle time

The QC process is often at the end of the Cycle in manufacturing environments (especially for the validation of finished goods). The product is invariably late and often unreasonable and unpredictable pressure is put on QC labs to reduce cycle times for testing. The effect is that the QC manager is unable to plan effectively. Lots arrive and are needed yesterday. The analysts are then unable to batch similar lots together to optimise set-ups and maximise output.

Also, customers are increasingly demanding smaller and smaller batches, and need to know the effects this has on costs. There is invariably no data to back-up the enforced inefficiencies that the lab has to face. There is a need for a tool to enable effective planning, to increase efficiency (by enabling batching etc.) and to help reduce cycle time.

Difficulties with planning & scheduling

The QC manager typically has to attempt to plan for and manage a very complex environment. It is essentially a production environment, so there are considerable volumes of entities and activities that have to be managed. However, it is made more difficult than many production environments by the complexity and variability of the activities carried out.

Typically in an analytical laboratory, there are a large number of tests, activities and 'other' tasks (training, cleaning, paperwork etc) that are very difficult to quantify. The activities and tests are very complex, involve a lot of sub-activities, and are very much up to the individual analyst to manage and optimise. The manager has little detailed information on how long these activities take, and the amount of resources they will take-up.

Usually in such an environment, the qualifications and capabilities of the individual analysts are very detailed and complex. An analyst may be qualified by customer, by product and by test (depending on customers, accreditation body etc.). This results in a very complex training matrix that can provide a headache even to the most intelligent planning system.

There is a need for a tool that can accurately model these activities and provide routings. Also a need to plan resources in the long term, medium term and schedule such complexity on a daily basis.

Tefen has applied standard industrial engineering practices to the problem, laid out in Fig. 2. The first question, 'Do I have enough resources?' is answered by creating a resource planning system. This is based in a database of test times constructed around groups of similar tests. The second question, 'How do I manage these resources?' is answered by a dispatching system. This uses an algorithm that matches tests in order of priority to the skills-matrix of the analysts. Both parts of the system are created and implemented as part of an efficiency improvement programme.

The model was developed as part of a Tefen effort to provide a tool to manage resources requirements corresponding to the manufacturing load, product mix changes, infrastructure for cost per batch analysis and initiate or support a process of productivity

resource planning module

The resource planning module is essentially a database and planning tool, which enables the following to be carried out:

The database is created using the following steps:

1. Using the list of products, their forecast, and estimated testing time requirements, perform a 80/20 Pareto analysis to identify the high running tests in order to focus data collection efforts.

2. Group tests according to work-centre (e.g. raw material, finished good, etc) similarity of activities, analyst activity time duration, and instrumentation.

3. The purpose of grouping the tests is to have a more manageable but still statistically representative number of samples that can be time studied in order to reduce the complexity and duration of this phase of the project.

4. Second Pareto Analysis – Extended detailed data collection and analysis will be conducted for the most common/frequent test groups (i.e. for the 20% of the groups that make up 80% of the total time per year). Fig. 4 shows an example of a Test Group Level Pareto analysis.

The dispatch module is essentially a scheduling system, which provides the following functions:

Dispatching system operation

Tefen's approach builds up the resource planning database as part of an efficiency improvement programme. The dispatching system is then built up on this basis.

A leading pharmaceutical company needed to measure the capacity of its chemistry lab in order to establish a planning tool to forecast optimal staffing and batching levels based on product mix and volume changes. Moreover, a scheduling tool was imperative in order to track cycle time measures and meet stringent due dates.

Tefen carried out an efficiency improvement programme that included the implementation of a resource planning and scheduling system. Extensive recommendations were made for activity automation, LIMS requirements definition, batch size optimisation, lot prioritisation and layout changes.

A number of improvements were seen as a result: