Evolving concepts in plant design: part 1

The construction of a new oral solid form (OSF) plant is an important decision and a real challenge. The team in charge of the basic conceptual design has to ensure that the new plant will be up-to-date and efficient not only at start-up, but for the next 15–20 years. This means that the project must be able to adjust to capacity changes, product changes and technology changes. It sometimes seems like an impossible challenge.

The basic conceptual design and selection of an appropriate plant layout cannot be improvized. It is the result of extensive experience, an in-depth knowledge of production processes and requirements, and the intelligent use of innovative concepts that have proven their value.

A good layout, based on a solid common sense, experience and innovation, will make the plant more cost-effective during its lifetime and will allow improvements to be made during the life of the plant without changing the basic design concept. Conversely, a bad layout will generate unnecessary costs, infringements to good manufacturing practices (GMPs) and leads to situations that are often impossible to correct after construction.

Often neglected, preliminary studies and production rationalization (before starting plant concept) are factors that generate significant savings, and have a dramatic impact on the final plant design. Our experience shows that rationalization of production and batch size optimization allows the reduction of the number of batches-per-year by a minimum of 50%. This reduction of manpower is linked to material handling, less sampling and analytical work, less cleaning and change over of equipment, less paperwork, etc. These aspects will be explained and illustrated in the second part of this article.

Conventional plant design

Model 1. Many OSF plants today are still mono-level buildings comprising a horizontal ground floor used for production and a light mezzanine above production used mainly for air conditioning.

In such plants, production rooms are usually accessed from one single main circulation corridor. In terms of people, this main corridor is used by all categories of operators: production, maintenance, cleaning, control, and often also by visitors. In addition, it is also used for the handling and transfer of raw materials, products in process, dirty and clean machine parts, garbage, etc. The schematic layout of such a plant is shown in Figure 1.

Figure 1 Model of conventional plant with only one corridor to access all production rooms.

Model 2. Many OSF plants in operation today use commercial containers called 'bins' or 'IBCs' (intermediate bulk containers) for the transport of products in bulk and for feeding production machines. In most cases, these containers are moved around by fork trucks or pallet movers that are running though the main circulation corridor and inside the clean rooms (Figure 2).

In some cases these engines are used in combination with lifting columns. This system should now be rejected for the supply of production machines because of cost and GMP reasons. Model 1 and Model 2 operations suffers from two major defects. From the GMP standpoint, the impossibility to reach a high enough level of cleanliness and the risk of mix-ups or cross-contamination. From a cost perspective, the wide corridors, the large surface areas needed for clean rooms and the important ceiling heights bring unnecessary expense.

Figure 2 Typical material transfer in conventional plant where bulk products are handled in clean areas and lifted above machines.

Model 3. In slightly better cases, still with mono-level plants, a second circulation corridor is provided at the back of the production rooms and is used for all so called 'dirty operations'. This is shown in Figure 3.

Figure 3 Model of conventional plant with clean corridor and a technical corridor to access production rooms.

No doubt this is an improvement: it provides flexibility for maintenance operations and allows the removal of garbage and dirty machine parts without contaminating the clean corridor. Unfortunately, it does not eliminate the risk of mix-ups within materials. Similarly it does not prevent fork trucks from circulating between clean rooms, nor does it eliminate the risk of cross-contamination.

Again, in all situations where IBCs have to be introduced inside the clean rooms, the use of lifting columns or of lifting trucks leads to the construction of clean zones about twice the size than needed and with a ceiling height 1 or 2 metres higher than required.

Model 4. Sometimes, use is made of flexible tubes, operating under vacuum, for interconnecting production machines. The use of vacuum implies at some point the use of cyclones and/or filters for separating air from product.

The cleaning of these systems and the need to replace the filters in case of product change represents an additional burden. In addition, if all machines are interconnected, the whole production line operates at the speed of the slowest machine and stopping any machine in the line halts the whole line. This probably explains why the vacuum transfer never did gather a lot of success.

Hub layout concept

It is must be underlined that efforts are constantly made to improve OSF plant concepts as illustrated by the hub layout concept.¹ Nevertheless, this concept is still based on IBCs circulating in clean rooms.

Consequently, GMPs are not optimized and this concept leads to unnecessary expenses for construction and operation. In addition the hub layout concept is applicable only for "manufacturing a number of similar class products on a campaign basis" which reduces flexibility.

The Lhoest concepts

In our opinion, one of the best ways to approach plant design is by implementing the concepts known in the professional literature as Lhoest Concepts. These internationally recognized concepts have been experienced, evaluated and improved since their inception. They are presently defined as the synergistic implementation of nine basic individual concepts, combined in a way that optimizes production conditions and maximizes flexibility, GMPs and cost efficiency.^2,3

The Lhoest concepts have been the basis for dozens of pharmaceutical manufacturing plants currently in use throughout the world. The first one was built in the 1980s in Spain and the most recent one, the Kowa plant in Nagoya (Japan) has just been nominated among the five finalists of the international competition for the 'Facility of the Year 2005' organized by the International Society for Pharmaceutical Engineering.^4,5

These pharmaceutical facilities have been approved by all authorities and have been successfully implemented in Belgium, Croatia, France, Germany, Holland, Italy, Japan, Norway, Puerto Rico, USA, Slovenia and Spain.

Figure 4 Separation of clean rooms (in blue) and technical areas (in clear grey).

The Lhoest concepts focus on nine specific areas:

Room segregation. One of the main concepts is the full segregation between clean production areas and technical areas (or optically clean areas), which means the relocation in technical areas of all material and pieces of equipment that are not absolutely needed in the production rooms (Figure 4). These include

• The materials feeding the process machines and intermediate bulk products that are stored in specially designed, tight stainless steel IBCs that are only handled in technical areas.

• Sources and distributions of energies.

• Pumps, gas cylinders, electrical motors, electrical boards, hydraulic drives and in general any component or machine part not absolutely required in the production rooms.

The installation 'through the wall' of production equipment is always recommended.

Containment. The transfer of products to and from IBCs has to take place under fully contained conditions. This not only renders the plants a lot cleaner but mainly avoids cross-contamination and any exposure of operators to active dust at any time. A key element is the connection system, usually called 'docking station', that may not allow the release of any trace of product at any time (during phases of connection, disconnection, standby, or during cleaning and inspection).⁶

The absence of any powder release applies to clean rooms as well as technical areas. Although a lot of suppliers propose connecting stations, very few connecting stations comply with containment according to Smepac rules.⁷ Connecting stations that do comply with Smepac have been tested successfully in Japan.⁸

The rule of containment should be applied throughout the whole production process.

Gravity transfer of goods. The transfer of goods, powders, granules, finished tablets, capsules, etc. by gravity is by far the simplest, safest and most economical one. For this purpose, the production floor is located between two technical floors (Figure 5).

Figure 5 The production floor is located between two technical floors.

The advantages resulting from the combination of the first three concepts are important from an economic and GMP perspective, namely:

• Significantly less ground space is needed because the plant is on three levels, sometimes five or more.

• Technical areas representing 60–80% of the total building volume can be constructed at about half the unit cost of clean areas (in a conventional plant, clean areas can, conversely, represent 80% of the total plant volume).

• The more compact construction allows shorter connections for all air and energy supplies. Additionally, because these distributions are located in technical areas, their unit installation cost is also reduced.

• Highly improved GMPs result from smaller and significantly simpler production rooms combined with full containment and elimination of IBCs and trucks from the clean areas.

Figure 6 Typical production island.

Islands of automation. In this concept, every individual production machine is located in a small clean room totally independent of the other islands. It possesses its independent air handling system, its own supplies and is able to run on its own at the optimum speed, just as a small independent production plant (Figures 6–7). The advantages are:

• Maximized flexibility — products can be transferred from one island to any other island in the factory. Changes in manufacturing processes or in delivery requirements have no impact on the plant Itself.

• A machine breakdown has minimum influence — the operation can be easily and rapidly deviated to another island. Maintenance is able to operate without any disruption of production.

• New pieces of equipment can be installed in spare rooms without influencing the system

• Every machine operates at its maximum speed.

• Automation allows some machines to run day and night, unattended.

Figure 7 Example of a typical production island: a calibration room.

It must be emphasized that applying these first four concepts permits the design of OSF facilities where different GMP classes of medicines (highly actives, antibiotics, etc.) are manufactured at the same time. Such plants have been approved by FDA. Islands of automation operate best in conjunction with the next five concepts.

Automation of material handling. The automated transportation of materials (powders, granules, tablets, capsules, pellets) to and from islands of production is cost effective. The automation approach must be adapted to the project and obviously manual transportation can also be used in countries with low labour costs.

All containers are standardized and equipped with patented connection devices to guarantee total containment and automatic, dust-free transfer under the most stringent conditions to agree with the new Smepac rules.

Depending on the case, the automated transportation is achieved by automated guided vehicles (AGVs, Figure 8), monorails, shuttles or elvecars (Figure 9). The experience has shown that in elevated labour-cost countries, this automation is always justified. The advantages provided by automated materials handling are:

• Experience shows that in a conventionnal plant, people use up to 40% of their time carrying or moving material around the plant. This time is fully recovered with automated transportation.

• The system operates unattended, 24 hours-a-day and 7 days-a-week, with resulting increase in plant capacity and efficiency.

• The materials handling is entirely located inside technical areas: clean rooms and clean corridors are no longer contaminated by the cumbersome handling of materials and can be built significantly smaller.

Table 1 Circuits of following classes of people and materials in plants have to be fully segregated.

Computer integration. With their impressive memory and their high level of repeatability, Computerized systems are an enormous help in tasks like management of inventories, storage of procedures, and specifications and assistance to manufacturing.

Warehouse integration. Considerable advantages result from the integration of the warehouse in the bulk production facilities:

• Exchanges of products between warehouse and production are implemented within seconds.

• Intermediary staging areas are eliminated with corresponding savings in manpower, surface area and paperwork.

• The inventory of the warehouse is updated automatically and in real-time.

• If the warehouse is constructed as the core of the production building, additional structure and isolation savings are produced.

Figure 8 AGV carrying a stainless steel IBC to a docking station.

Cleaning-in-place of machines, containers and installations. This concept is well known. It is applied whenever justified.

Segregation of circulations. Fully segregated circulations are provided for the categories of the people and materials shown in Table 1 that are practically always present in any OSF plant.

It is essential, under the Lhoest concepts, to ensure that no cross-contamination can possibly occur through the circulation of different classes of people or materials.

In other words, dedicated changing rooms, provided for each category of personnel, lead through dedicated corridors, directly to the rooms authorized to that particular category, and without any possible interference from other classes. The same applies to materials. This objective is not easy to achieve, but the best OSF plants in operation today are based on this important principle.

Figure 9 Elvecars run twice as fast than AGVs and do not require any special flooring.

Conclusions

We believe that implementing these nine concepts is the best approach to design a cost-effective plant that will meet all GMP requirements and remain extremely flexible.

Not all of the nine concepts need to be implemented; each project has its specific requirements, but to get the necessary synergy, at least six or seven of these concepts must be implemented. Although individually taken, each concept looks easy to apply, their combination requires a lot of attention, coordination and experience.

Figure 10 The automated high bay warehouse as the core of the bulk production plant. A seperate warehouse located at a distance from the main production building can be a serious handicap and always causes delays, staging, and additional paperwork. In an OSF plant with some degree of automation, it has been found advantageous to locate the raw materials warehouse, or at least a buffer warehouse, adjacent to bulk production. Alternatively, the bulk production can be built around the raw materials warehouse. It becomes thereby an integral part of the general materials handling system.

In Part 2, case studies of Lhoest type plants in operation will be discussed and compared with conventional plants.

Jacqueline Vu is senior process engineer in the pharma business unit of Coppée-Courtoy, Belgium.

References

1. M.P. Brocklebank, J. Lam and P. Mehta, Pharm. Engineering.26(1), 68-78 (2006).

2. M. J. Cliff. European Pharmaceutical Review 25–35 (June 1996).

3. G. Cole, Pharmaceutical Production Facilities. Design and Applications. (Ellis Horwood, Chichester, UK, 1990), 69.

4. W. Lhoest and P. Froment, Pharmaceutical Technology 28–40 (June1984).

5. Pharm. Engineering Special Edition pp 12 (April 2005).

6. W. Lhoest, Pharm. Engineering.22(2), 8–25, (2002).

7. ISPE good practice Guide — Assessing The Particulate Containment Performance Of Pharmaceutical Equipment. ISBN 1-931879-35–4 (2005).

8. H. Masuda et al., Pharm Tech Japan,21(6), (2005).