Companies unable to increase manufacturing flexibility are operating on borrowed time in today's climate. Filippo Focacci, ILOG Supply Chain Applications, and David Simchi-Levi, MIT professor and ILOG chief science officer, explain how it can be done
Despite the continuous growth of the pharma market, the industry is facing new challenges that are forcing a redesign of the supply chain and manufacturing practices. Typically, the industry has been characterised by very high r&d costs, high production costs (especially in biotechnology), heavy regulations, low operational efficiency, long cycle times, and high inventories. Yet, at the same time, the industry managed to sustain high profit margins so that operational efficiencies - either in the supply chain or in manufacturing - generated very little interest from company executives.
However, times are changing.
Economic changes, including pricing pressure from governments, counterfeiting and competition from generics as well as a shortening of patent protection times and scientific-technical evolutions, are creating new challenges that will have a deep impact on the pharmaceutical business. The industry is facing an increase in complexity and variability and, as a result, new supply chain and operating models are urgently required.
To overcome these new challenges, the pharmaceutical sector must look to other industries and learn from their best practices in and successful use of Lean Manufacturing to improve their operations. Not surprisingly, the most successful lean implementations are found in the discrete industries characterised by high volume, repetitive operations and assembly lines: lean is an excellent fit for overcoming many of the challenges faced by these companies.
In the past few years, AMR Research has been promoting Demand Driven Manufacturing and Demand Driven Supply Chain best practices, which focus on better synchronisation between manufacturing execution, supply chain processes and demand sensing.
key objective
Improving flexibility is a key objective of Lean Manufacturing and Demand Driven Manufacturing best practices. It is our opinion that business processes and technology that enable companies to be more flexible and agile will be important in ensuring companies remain profitable and grow in the face of increasing complexity.
The main objective of Lean Manufacturing is to provide the best possible service to customers through the elimination of waste. Waste can take the form of material or energy waste, inventory, defects or wasted capacity. To avoid waste, manufacturing execution must be tightly synchronised with supply chain plans and customer orders so that throughput of the manufacturing process is equal to customer demand.
Lean techniques have been driven by the requirements of assembly line manufacturing. Achieving similar results in different manufacturing environments requires an understanding of the key lean principles:
Take a holistic view. The production environment needs to be optimised as a whole with the aim of closely co-ordinating operations that are physically separated.
Attention to detail. Operational details are strategically important. The focus on set-up reduction is a good example. Instead of taking set-ups as a fixed constraint of the system, engineers try to reduce the set-up time so that non-productive operations are minimised.
Control work in progress (WIP). This is an important objective achieved through the use of Kanban cards. The key idea is to change the manufacturing process from a "push-based" manufacturing process to a "pull-based" one.
Reduce cycle time. This key objective is achieved by reducing set-ups and delays, co-ordinating machine maintenance with production operations and optimising space to better utilise workers and equipment.
Implementing lean techniques is not merely challenging in the pharma sector, but can generate poor manufacturing performance if not implemented correctly. For example, reducing lot sizes in this industry may lead to waste through low utilisation of tank capacity.
More generally, it is not clear how to implement a pull strategy in a production process characterised by strong economies of scale and/or batching. If raw materials and intermediate products are produced and stored in tanks, it becomes very difficult to implement pull scheduling and a one-piece flow process (where each product moves through the process one unit at a time). Similarly, the Kanban concept is appropriate for discrete manufacturing, but hard to implement in pharma production.
Furthermore, quick changeovers can be difficult, or impossible when cleaning-in-place (CIP) may require a full day.
lean principles
Although traditional lean techniques cannot be applied to the industry, lean principles can be applied to identify requirements for implementing effective manufacturing strategies in the pharmaceutical industry:
Take a holistic view. Sequencing and scheduling decisions must be made by looking at their impact on both resource efficiency and inventory levels, and including information about raw materials, intermediates and finished goods.
Attention to detail. Operational details must be taken into account when generating production plans and schedules. Building a plan that ignores key manufacturing constraints such as tank capacities, cleaning rules or sequence-dependent changeover times will generate continuous adjustments to the plan. Additional changeovers generate loss of capacity and inventory shortages, which in turn generate expedite orders and new infeasible plans.
Control WIP. WIP can be reduced by carefully co-ordinating intermediate product manufacture, finished good production and demand signal.
Reduce cycle time. Cycle time includes two components: wait time and processing time. Wait time is reduced by better co-ordinating the flow of material (e.g. intermediate products and finished goods), while processing time reduction is achieved through carefully balancing changeover times and costs with inventory costs.
These requirements suggest that there is a need for effective optimisation methods that take a holistic view of the production process and balance the various trade-offs while considering all business constraints. Such methods also provide planners with enough flexibility to easily and effectively modify production plans.
The revenue of biotech companies in the past 10-15 years has experienced exponential growth, but profits have not always met expectations.1 In fact, r&d costs are extremely high, building biotechnology facilities is very expensive and these facilities are often inflexible as they tend to be designed for a single product.
Manufacturing processes in biotech are extremely complex, involving cell growth in bioreactors, a sequence of purification steps in chromatography columns and filtering equipment, but also involving the manufacturing of media and buffer preparations necessary in different processing steps. All these steps must be tightly co-ordinated and involve shared resources, such as labour and equipment for CIP. In the past few years there has been a recognition that both manufacturing operations and plant and process design are key to profitability in this industry.
high cost of change
Plant and process design is particularly important in biotech where building a plant may take up to eight years and cost billions of pounds. Even a change to an existing plant or process may be very expensive and time-consuming. In fact, the FDA may require companies to re-do clinical trials for changes in existing processes. Therefore, decisions concerning investments in new facilities or expansion plans and competing biotech technology (e.g. continuous perfusion vs batch) have an impact on long-term profitability. Companies need to be sure that these decisions will meet requirements of unknown, long-term outcomes.
Biotech companies today employ large process engineering groups to analyse and optimise manufacturing processes. Most process engineers use simulation tools that enable them to develop models for mass energy balances, estimating waste and peak and average utility consumption etc.
While these tools enable engineers to model chemical and biological processes precisely, they lack the sophisticated optimisation capabilities necessary to provide a holistic view of the manufacturing processes and to generate realistic plans and schedules taking into account all manufacturing constraints. Using simulation tools, engineers are able to simulate the dynamics of a single batch in an empty plant, but they are unable to simulate the intricacies and the dynamics of several batches running on the same facility and using shared equipment, support operations and utilities.
Today, the main goal of plant and process design is to find the right size of equipment and operating procedures for new facilities, expansion plans and process changes. In this context, scheduling of batches becomes more important than generating a very precise model of chemical and biological reactions.
ILOG Plant PowerOps is used by industrial engineers to simulate different plant configurations using a simple user interface. Designed for manufacturing planning and scheduling, it enables different configurations to be tested against alternative future demand trends to balance costs and manufacturing capacity. It uncovers hidden bottlenecks and facilitates strategic decisions based on realistic simulations of manufacturing schedules.
The ability to use the same tool for plant design and production planning and scheduling has additional advantages. It allows planners and engineers to work off the same data and make decisions based on the same manufacturing models.
To cope with increasing complexity and variability, pharmaceutical companies need to improve processes and operations so that they become more flexible. While statistical analysis can help companies reduce variability, optimisation is often at the heart of decision support tools helping companies increase flexibility. Integrated planning and scheduling tools are key to manufacturing flexibility which has to be the priority for today's companies if they wish to compete in this changing industry environment.
Results and impact on flexibility when using ILOG Plant PowerOps
ILOG's planning and scheduling solution Plant PowerOps (PPO) was implemented and the results analysed in terms of inventory coverage, 'production smoothing' and operational efficiency. The results demonstrated a reduction in inventory excesses (almost down to zero), and deficits. ILOG PPO was able to achieve this by finding the right balance between inventory costs and changeover costs under tight resource capacity constraints.
The results also illustrated a significant increase in operational efficiency, which was improved by between 2% and 5%. The total time due to non-productive activities, such as CIP and changeover, was also reduced by 10 â ‚¬ Euro Å" 40%.
Obviously, the reduction in changeovers and cleaning and the increase in operational efficiency are linked. In fact, operational efficiency is defined as the ratio between the operational time and the net production time and the reduction of changeovers and cleaning implies a reduction of operational time without impacting the net production time.
'Production smoothing' was also improved: the production frequency of the optimised plan was significantly more stable than the frequency of the manual plan and the production variability (in volume) was also reduced.
Among the many benefits of this implementation, the two that directly impact manufacturing flexibility are: increased operational efficiency and faster rescheduling.
Increasing operational efficiency means that the plant is able to produce the same output using resources for a shorter amount of time, thus creating a 'buffer' of resource capacity. These capacity buffers create manufacturing flexibility and can be used in many ways:
- To increase volume without making heavy capital investment
- To accelerate product launches and react to market changes
- To reduce the impact of production variability.
The ability to increase throughput without making changes to the production facility or to the production process is particularly important in the biotech industry, where equipment is costly, changes in production may have to go through FDA approval, and product stock-outs are often unacceptable.
It is also easy to see how reducing planning and scheduling cycle times from 1-2 days to a few hours directly affects manufacturing flexibility. If planners are able to generate a feasible plan in a few hours, they will be able to better respond to market changes or business requirements and to react better to production problems, e.g. if a batch is segregated.