Manufacturing guide wires is a very intricate operation involving a variety of both manual and automated tasks. Despite the heavy ISO and FDA focus on procedures, the majority of medical device recalls can be traced back to a failure to follow procedure. Therefore, within the manufacturing of medical devices, a company’s adherence to ‘standard work’ is critical.
The concept of standard work goes back as far as its first major advocate, Henry Ford, who said:
There is only one best way. What is the best way? It is the sum of all the good ways we have discovered up to the present. It therefore becomes the standard. Today’s best, which superseded yesterday’s will be superseded by tomorrow’s best
Ford’s philosophy is just as relevant today as it was over 100 years ago. The design and implementation of standard work for medical device production has proven a major challenge. It forces management and engineering staff to clearly identify and define the one best way. Standard operating procedures (SOPs) have tended to be voluminous documents that are difficult to understand and are open to interpretation. Standard work is very definitive – it leaves no room for error and forms the basis for understanding and learning.
There is also a people paradox in managing improvements and change in the medical device sector. While today’s production associates are very well educated and highly skilled, they tend towards innovation as a means to achieve improvement. Whilst this is much sought-after in employees, it must be controlled or it may become a source of inefficiency and the antithesis of improved performance that standard work enables. As well as ensuring that every single product is manufactured fit for its intended use, standard work also acts as the baseline for all future improvements. Without standardisation there can be no stabilisation – and without stabilisation there can be no maximisation.
Phase 1: Review of Standard Operating Procedures
A review found that the quality of the SOP documentation was of the highest level and the systems in place to manage and record employee training were extremely effective. However while all these systems were adhering to best practice in FDA regulated industries, three major opportunities for strengthening the systems were identified:
- Experienced production associates had a hidden layer of detailed tacit knowledge which was not shared or captured anywhere. In fact this tacit knowledge was key in getting to the root cause.
- The majority of SOPs were written in engineering speak and very broad, e.g. detailing equipment settings, operation, asset care, maintenance, etc., which resulted in a typical SOP being >30 pages long (with some even as large as >100 pages). This called into question their suitability as a training aid for new employees.
- While all production associates had been trained and were certified to carry out their duties as prescribed in the SOPs, in reality they had no real or deep understanding about the tasks they were performing. Therefore, they did not understand the reasons why certain critical tasks had to be performed in a particular way.
Phase 2: Introduce Job Breakdown Sheets (JBS) in line with Training Within Industry (TWI) Philosophy
TWI differs from traditional training systems in one fundamental aspect. Rather than just creating a set of instructions and training the associate to follow them; it embeds a real understanding of the process. This is achieved by teaching them, in a layered manner, the significance of each key step in the process and the consequences if these key steps are not performed correctly. Note: A “key step” is one which can adversely affect quality, safety, productivity, or cost.
A pilot area was selected to conduct a controlled trial. The 35 page long SOP was refined down to three specific JBS – one detailing operation, one for asset care, and one for set-up – which were developed by the CI engineer with input from the relevant operations, quality, and maintenance personnel.
There were four key steps in developing the JBS:
- Go to Gemba (factory floor) and involve the associates at every stage of the project.
- Watch every motion in great detail, using video analysis where required.
- Use data to determine the current one best method.
- Capture this method in non-technical language.
At this stage Quality Assurance was invited to review the three documents to ensure that they complied with ISO 14001 and FDA requirements as per the company’s quality management system. To trial the system, four production associates were selected for training (two using the traditional method and two using the TWI method). This provided an opportunity to conduct an experiment to compare results using the standard training method against using the TWI method. The results were very dramatic – see the graphs below.
Based on these extremely positive results the decision was made to restructure Lake Region’s training in line with TWI philosophy. This resulted in a complete review of all production SOPs and the creation of over 100 job breakdown sheets. These JBS were laminated and installed where they could be of most benefit to the production associates, i.e. at their work stations. Likewise, they were available for team leaders to pick up and observe immediately whether standard work was being performed correctly.
It is important to reiterate that the main purpose of these JBS is to assist in the training of manufacturing personnel and to instill understanding and knowledge of the processes.
There is still a requirement for the detailed SOP, and in fact every production line is equipped with a PC to enable the production associates access broader details, e.g. equipment settings for non-standard parts, troubleshooting guides, etc.
Phase 3: Introducing Standard Work Compliance Checks
In order to support the roll-out the implementation team developed an auditing system to measure how well standard work was being followed. Each week an audit team consisting of a CI engineer, production manager and training specialist carry out a process compliance check on a production or packaging cell which is chosen at random. Each audit team member (using the relevant JBS) observes a different production associate performing their work and records the result on the sheet. While the audit is being carried out, the line JBS are reversed to prevent the operators using them.
When the check is completed the audit team collates their final scores, gives feedback, and circulates the results via e-mail to the key stakeholders. During a typical check which lasts between 45 to 50 minutes, between 250 to 300 tasks can be checked. One key refinement that the team has made to the check is that in the event of a production associate either missing a critical quality or safety step, or performing a critical quality or safety step incorrectly, production is immediately stopped until the issue is resolved and 10% is deducted from the final score. This supports the company’s key objectives of supplying guide wires of the highest quality while ensuring the health and safety of their employees. As well as auditing how work is being performed, this check also presents an opportunity to assess the quality of the JBS and look for opportunities to improve the process. Input from operations personnel was key in the creation and introduction of these standard work compliance checks. In particular the team wanted to assure the production associates that the focus of the compliance check was on the process rather than the person. In the event that an associate is observed not performing a key point correctly he/she will be advised as to the correct method. This information will be passed on to the relevant team leader who will first organize for the associate to be retrained and will then audit them daily until he/she is satisfied that the associate’s performance is 100% satisfactory.
The figure below outlines how the auditing system is designed, not alone to check if the production associates are performing the task correctly, but also to ensure that the work instruction is adequate and that the training has been delivered correctly.
This is a synopsis of an article authored by Noel Hennessey which appeared in the Lean Management Journal in September 2012. (Not for reproduction with the prior authorisation of the author.)