In this session, Prasad Saraph from Bayer Healthcare’s Biotech Pharmaceuticals Division provided a fascinating insight into the supply chain issues that Bayer faces in producing one of its key biotech pharmaceutical products.

Kogenate is a leading treatment for hemophilia-A, a genetically transmitted condition that prevents blood clotting and affects 400,000+ people worldwide.

Prasad began with some startling figures: 1,600 employees in 15 manufacturing plants work continuously throughout the year to produce just 250 grams of active ingredient, supporting over one billion USD of annual sales of Kogenate.

250g. $1bn. Wow.

Biotech pharmaceutical companies function in massively regulated environments. Add to that the technical complexity of the supply chain and manufacturing process and you can understand why companies like Bayer can leave nothing to chance.

All batches follow the same process flow but may use alternative facilities at each stage: Expression, fermentation, purification, bulk, fill, freeze dry, and packaging. Only one batch can be processed in a facility at a time (to avoid cross-contamination). In certain stages of the manufacturing process, employees must work in multi-layer suits, with highly advanced air filtering and evacuation processes and for no longer than four-hour shifts at a time.

Demand side cosntraints: Large demand fluctuations. The tender process which is increasingly being used by customers, and public authorities creates uncertainty (you’re either in or you’re out of a market, depending on whether you win or lose the tender). Customer emergencies (accidents, cuts) can cause huge demand increases compared to normal prescription levels. Inventory risks: Cold chain needs to be permanently maintained (product must be kept refrigerated 2-8°C) - suppliers must always maintain mandated inventory levels.

Supply side constraints: Complexity in the process, regulatory approval, equipment, facility, and formulation. Process uncertainty: varying quality cycle times based on deviations. Varying reject rates. Biotech products are made using live proteins. The compounds are not produced synthetically and therefore the whole process is less predictable.

Supply chain Planning is a continuous process, with endless cycles of data updates, assumption updates, scenario revisions.

Bayer has close ties with UC Berkeley and is highly involved with the Biotech Forum.

Key objectives: Produce sufficient but not excessive material, as soon as possible, but not too soon, while meeting all regional regulatory requirements on product, process equipment, facilities and raw materials.

Key considerations: Traceability, No back orders. Stock outages is not an option (= loss of life … and law suits).

Bayer uses CPLEX for MIP formulation because it can handle their large problem size. It’s a reliable and proven platform. They needed a platform that developers were familiar with.

Solution times: 30,000 boolean variables, 500,000 float variables, 100,000 contraints. 90% optimum solution reached in about 12 hours on a 1.8Ghz quad core dual processor 64 bit machine with 16Gb RAM.

Challenges: Data quality from ERP, planning engine user interface bugs, interpretation of PE outputs, keeping PE assumptions up to date, WIFM issues from key users and end users

Benefits: One version of the truth. Data fidelity, scenario management, mid-term planning decision support (what-if scenarios). Planners are no longer afraid of calls from regulatory bodies. They can react much more quickly and easily.

The future: Ability to analyze historical batch data to better improve planning forecasts.

Posted by Stuart Culshaw in Dialog 08, Dialog day 3, Supply Chain Management track. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.