How do you maintain lot integrity in a continuous manufacturing process?

As continuous manufacturing (CM) becomes more prevalent, the question of lot integrity will become more important.  The company where I work presently has been dealing with this question for at least as many years as I have been here and cannot seem to come to a consensus of an acceptable path forward.  In previous lives at companies that had implemented CM, there was an "assumed" amount of carryover based on theoretical calculations of purge volume from transfer tank to filling head.  This may not have been a full CM process in that bulk was still being manufactured by individual compounded batches.  However, even in a truly full CM process, the addition of new raw materials constitutes a new lot.  How does one identify this lot as distinct from the previous?  And at what point does the new "batch" of materials constitute a new finished product lot at the filler/finished package?

Parents
  • Dear Jeffrey, 

    our sincere apology for the delayed reply. No excuse but we had just a alignment internally about our answer as your question is indeed more complex than to answer in a few sentences. We also assume, as you refer to a "transfer tank" and a "filling head", that you are working on a CM liquid process? Before providing real numbers and detailed answers we would need to know some more details about the CM system you are working on. 1.) On OSD, powder based CM systems the track & trace is based on RTD models which are developed based on the materials and capacities involved. Each process system has its own RTD model depending on the unit operations and size etc. The so called v0 or free volume or volume at t=0 is a constant which will be calculated by the equipment vendor or system integrator. But v0 is crucial together with the RTD to calculate the back mixing and the start-up or ramp-down volume etc. Also when it comes to the cleaning and emptying of the system this is crucial to know. 2.) It is important to understand the inbound material identification. On most of full CM system is a NIR system applied to ID check the inbound material. 3.) Usually the first unit op on a CM system is either a "loss in weight" feeder or liquid dosing system based on "loss in weight" or flow control measures. Those elements, together with the ID check, are embedded PAT tools which will detect any changes in the consistency of a inbound material (e.g. fluctuation of bulk density? which will indicate immediately and real time, any change of different lot. All together this plays into the over arching control strategy where the so called batch & lot genealogy is managed. The other approach is based on a pre-defined material balance. But this is usually only for very small batch sizes important where no automated material refilling is required. 

    Sorry for this long answer but I am afraid for a real comprehensive reply we would need a F2F meeting, or at least a telcon. To support clients on those developments is one of our key PCM services. 

    I am looking very much forward hearing back from you. 

    Richard     

Reply
  • Dear Jeffrey, 

    our sincere apology for the delayed reply. No excuse but we had just a alignment internally about our answer as your question is indeed more complex than to answer in a few sentences. We also assume, as you refer to a "transfer tank" and a "filling head", that you are working on a CM liquid process? Before providing real numbers and detailed answers we would need to know some more details about the CM system you are working on. 1.) On OSD, powder based CM systems the track & trace is based on RTD models which are developed based on the materials and capacities involved. Each process system has its own RTD model depending on the unit operations and size etc. The so called v0 or free volume or volume at t=0 is a constant which will be calculated by the equipment vendor or system integrator. But v0 is crucial together with the RTD to calculate the back mixing and the start-up or ramp-down volume etc. Also when it comes to the cleaning and emptying of the system this is crucial to know. 2.) It is important to understand the inbound material identification. On most of full CM system is a NIR system applied to ID check the inbound material. 3.) Usually the first unit op on a CM system is either a "loss in weight" feeder or liquid dosing system based on "loss in weight" or flow control measures. Those elements, together with the ID check, are embedded PAT tools which will detect any changes in the consistency of a inbound material (e.g. fluctuation of bulk density? which will indicate immediately and real time, any change of different lot. All together this plays into the over arching control strategy where the so called batch & lot genealogy is managed. The other approach is based on a pre-defined material balance. But this is usually only for very small batch sizes important where no automated material refilling is required. 

    Sorry for this long answer but I am afraid for a real comprehensive reply we would need a F2F meeting, or at least a telcon. To support clients on those developments is one of our key PCM services. 

    I am looking very much forward hearing back from you. 

    Richard     

Children
No Data