Rise of the robots

Not all manufacturers are taking full advantage of advances in robotics to improve efficiency. I have been to countless pharmaceutical plants in my time as an engineer. In all of them, the single most important factor is the cost of the end product. Naturally, the machines used to produce that, end product must be appropriately validated — and the process itself must be sufficiently traceable to comply with all relevant regulations.

Regulatory compliance

Of the numerous regulations that are applied to the European pharmaceutical industry, two are particularly important: 21 CFR Parts 11 and 211 — both FDA regulations. However, 21 CFR Part 11 has become the de facto world pharmaceutical industry standard. European countries have their own rules and regulations for the production of pharmaceuticals, but the pervasive footprint of the US in the global market place has made the US standard a global language.

21 CFR Part 11 relates to guidelines for trustworthy electronic records. This regulation requires companies to use procedures and controls that are there to ensure the authenticity, integrity and confidentiality of electronic records. In addition, 21 CFR Part 11 helps certify that an electronic signature is genuine. Part 211 spells out the minimum current good manufacturing practice (cGMP) for the preparation of drug products. This includes aseptic processing involving any robotic equipment. Equipment used for pharmaceutical production must be routinely calibrated, inspected, or checked according to a written programme designed to ensure proper performance. Whilst compliance is a requirement for all pharmaceutical manufacturing processes, whether they incorporate robotics or not, there are ways that robots can help to achieve compliance.

For example, cGMP is fundamentally a list of rules on how to create your own rules, not a list of rules of what you can and can't do. While the regulations do not require the use of robotics, the regulations state that if a company is going to deploy robots as a means of manufacturing pharmaceuticals, there are standards to which they must be maintained, cleaned, inspected and calibrated.

These standards are met using a plan drawn up by the company itself. The role of FDA is to examine and approve the company's plan and to ensure that it is being followed faithfully. As a result, ironically, it's the people writing these procedures that make them stringent — so they are more likely to be approved by FDA. The automation designer or systems integrator has to follow these procedures which involve design concepts, design reviews and peer reviews. If there is a change, the company has to have a change record, which must also be approved. However, because the company plans these procedures itself, there is a danger of over-elaboration.

Key points

Nevertheless, the requirement for electronic records is a perfect match with robotics. It is a relatively simple part of work cell integration to have the robot record each step it undertakes so that data can be be tracked down later. Furthermore, the data can be fed to a programmable logic controller (PLC), human machine interface (HMI) or supervisory control and data acquisition (SCADA) system to be analysed should there be a recall of the product by FDA or if the process is being streamlined.

Another facet of 21 CFR Part 11 states that there should be an audit trail when it comes to generating data. These audit trails must be secure, computer-generated and time-stamped to independently record the date and time of entries. Any activity that creates, modifies, or deletes electronic records has to be stored and made available should there be a need for FDA to conduct an audit. Audit trails are required to specify who did what to the records and when this was done. If the robot cell is being used to control other elements of the plant's equipment, which is possible where the robot controller incorporates a built-in PLC for example, then this can be achieved easily.

The date the data was collected, the conditions it was collected under, and who collected it need to be included, which are requirements complementary to the use of robotics.

Cost-cutting exercises

I've seen lots of plants take the simplest possible approach to reducing prices — they don't spend a great deal of money on technology in the first place. Clearly, pharmaceutical manufacturing is not in the dark ages, inspection in particular is highly sophisticated. However, one area where spending less doesn't necessarily lead to lower costs is manufacturing efficiency. I have seen manufacturing plants that were very cheap to set up and represent extraordinary examples of lean manufacturing theory in practice. However, these same plants often fail to invest in machinery that could increase output and lower overheads because of the initial capital investment. Robotics can increase efficiency, which means the price of the drug itself will become more competitive.

However, to obtain this quick return on investment (ROI) and comply with all relevant regulations, one must specify the right system in the first place. When doing so, the key factors to address are speed, payload and flexibility.

The speed, or cycle time, is an essential consideration in pharmaceutical manufacturing. For example, a broken ampoule or spilt syrup can mean a breach of the aseptic environment which leads to costly down time and significant financial losses. Cycle times are usually less than four seconds in pharmaceutical applications.

Another significant factor is payload. In pharmaceutical manufacturing or packaging applications, robots can often be required to lift heavy items. However, payload is not a factor to be ignored in laboratories. It is worth considering the inevitable trade-off in repeatability and accuracy that is required if payload is to be increased. It is a good idea not to over-specify payload 'just in case' because an increase in payload means reduced repeatability.

Types of robots on the market

The third of the three key issues to consider is flexibility. I have often encountered applications where selectively compliant articulated robot arm (SCARA), Cartesian or even six-axis robots could perform effectively and meet the requirements.

This illustrates the flexibility of the different kinds of robotic systems on the market (see sidebar) — the same machine can find different uses in packaging, laboratories and manufacturing. However, key things to consider are the ease with which the robot can be programmed and the space it occupies. The most important factor, of course, is programming simplicity.

If, for example, a single manufacturing line in a contract manufacturing plant is producing syrups for one client for the first half of the year and another client for the second half of the year, the re-programming of the robot should be as simple as possible to make the changeover easy. Simple programming languages such as SCOL (a hybrid of the multi-platform computer programming language Java and VRML that easily enables the creation of 3D sites) teach pendant functionality (through which the robot is 'taught' its movements) and 3D simulation software can all be benefits in this context.

With regard to size, choosing a robot for its flexibility is a very simple issue. The robot with the most payload, speed and functionality in the smallest possible footprint is the one that is likely to be most useful in other applications later in its lifetime. This size consideration has led to a trend towards smaller robots — decreasing the footprint within pharmaceutical work cells. However, there are other benefits that should be considered. For example, SCARA robots can be quite tall and the wiring harness adds to this height. However, if it's built into the machine, the robot can be repurposed for use in smaller spaces.

Most SCARA robots used by pharmaceutical companies are used on packaging lines. Like other products, pharmaceuticals need to be packaged and inspected after production. Often, medication is packaged in blister packs to protect cleanliness and to help patients keep track of their daily dosages. This kind of packaging is repetitive but could also demand that the purpose of a packaging line is changed frequently. The flexibility and easy programming provided by SCARA robots is particularly suited here.

For example, blister pack packaging is pretty fast, but not as fast as bottling. The speeds have to be kept relatively low, about 30/min, to prevent tablets from breaking. More robust products can be packaged at 30–60/min, while big selling, commodity, consumer generics can go as fast as 200 pills/min. This is well within the capability of a good SCARA robot.

Robots now also play an essential role in the development of new drugs. In high throughput screening (HTS) for instance, millions of compounds are tested to determine which could become new drugs. The use of robotics can speed this process up significantly, just as they can any other process where a robot replaces a person completing any repetitive task. Of course, unlike manufacturing, the profit centre in this context lies in speed to market, so capital investment in automation is more common.

Prior to the widespread use of robotics in the drug discovery process, pharmaceutical companies screened compounds manually, literally pipetting them by hand. It could take years to sample enough compounds to find a good lead with a typical lab only able to screen thirty or so compounds a week. Using automation, there is the potential for a single robot to screen hundreds of compounds in a twenty-four hour cycle. As a result, researchers can now screen an entire sample collection in approximately three months.

Once the compound has been developed, SCARA robots can find applications in the process of developing the compound further. A range of robot types are used in creating pharmaceutical libraries as well as for laboratory testing — things like testing blood and urine samples are common applications.

Another advantage in the laboratory is that robots are impervious to many environments that would not be safe for humans. A robot can operate twenty-four hours a day, seven days a week without a dip in accuracy or production.

When it comes to pharmaceutical production, people are not as efficient as robots, especially when they are wearing a protective suit. People in protective suits also require more room to work in.

This ability to work in hostile environments means that SCARA robots are used in a range of specific applications. For example, some experiments require temperature to be controlled exactly with the robot moving samples in an out of an oven for a specified incubation period. This is the perfect application for a SCARA.

Robots can be used in a wide variety of applications in the pharmaceutical industry and their role in manufacturing is definitely on the rise.

Nigel Smith is the managing director of TM Robotics, UK.