What We Delivered

We agreed with the client that powder material variables would be the best place to start our investigation. Key areas for potential variable change included:

• sourcing raw material powder
• storage in silos
• and metering out (on a batch to batch basis).

To make things even more complicated the materials were bought in and consumed at different rates, meaning overlaps in batches of different materials.

A common challenge for manufacturers needing to perform investigations is that they also need to keep production going with minimal or no disruption. We suggested utilizing retrospective Factorial Experimental Design (FED). Using this technique we were able to use existing data (in this case particle size distribution (PSD) of raw material powders) instead of requiring a matrix of controlled experiments that would require the use of materials and processing time.

PSD represented our variables while final tablet strength and breakage percentage were taken as our responses, so what were we able to learn?

We were able to determine that the cause of the problem was two raw materials in particular. Issues with batch inconsistency (excessive coarse and fines in some batches) were strongly correlated with poor tablet strength and breakages.

Armed with this knowledge our client was able to work with suppliers to determine a tighter tolerance for their raw materials' PSD enabling significant reduction in breakages and waste to be achieved.

Value to the Client

Our analysis showed the variation in raw materials to be the primary factor for our client's challenge, but always keen to go above and beyond, our experts also noticed correlations with other variables, including:

• shift times
• shift teams
• and specific tablet presses being used.

These observations formed the basis for a second phase of FED investigations to help reduce processing issues even further.