Minimizing Cost of Product Transitions
With more than 500 projects completed with less than one year to payback, Cargill Optimizing Services understands how to decrease costs and increase yields in the food and beverage industry.
Many food and beverage plants produce multiple products using the same equipment and transfer lines. The strategy used to make this transition can have a significant impact on the energy and water use, and on the overall yield and waste produced. However, it is often difficult to quantify the impact of these transitional costs, so the production teams often lack the data to influence the number of product transitions proposed by the scheduling or sales teams or to persuade quality or food safety teams to change the protocols for product switching.
This article highlights how dynamic simulation can be used to quantify the costs of a transition and to review the impact of various possible changes on product purity. This information can then be used to justify carrying more inventory or to accept a greater degree of product co-mingling.
- Asset Utilization
This analysis demonstrates the use of simulation to elucidate product transition strategies. Transition costs are difficult to quantify, and often lead to conservative strategies being imposed by, for example, Quality Assurance. It can also lead Inventory Management teams to request multiple, short batches that minimize Work In Progress. Understanding the cost implications of these decisions can allow significant cost savings to accrue, as well as waste reduction and asset utilization improvements to be achieved.
In this analysis, we review a product running through two pipe runs separated by a valve cluster and filling one of two tanks. We model running with one product, flushing the lines, and then sending another product to the second tank, filling it to 10%. The products contain water, sugar and two flavor components, and the cost of the transfer is calculated considering the raw material and the waste disposal costs – other costs could be considered, but are neglected for simplicity. The cost and composition is summarized below.
|Component||Cost||Product 1||Product 2|
|Flavor A||40.0 $/kg||150 ppm||75 ppm|
|Flavor B||20.0 $/kg||50 ppm||200 ppm|
|Sugar||0.5 $/kg||29.99 wt%||29.99 wt%|
|Water||2.0 $/m3||70 wt%||70 wt%|
The base case strategy involves flushing the lines with water between the two product runs. As soon as water is introduced, the material leaving the transfer lines is diverted to the drain and the lines are flushed for 2 residence times (by which time the solution is ~99% water). The water is then stopped and Product 2 is introduced – material flows to drain for 3 residence times (by which time the material is within 1% of target). The concentration of material leaving the transfer pipe is shown below. As can be seen in the Base Case of Table 1, this ensures that the material in the two tanks is of almost identical composition to the target. It is, however, a relatively expensive strategy – each transition of this type costs over $300.
If some composition deviation is acceptable, then a number of strategies can reduce cost. For example, some of the initial “waste” material (diluted Prod-uct 1) can be recovered. Similarly, as Product B pushes out the wash water, some of the transition “tail” can be recovered. The simulation was used to identify conditions that kept the final composition of material in the tanks with-in 1% of the specification. As can be seen in Case 1 of Table 1, accepting this deviation from spec reduces the waste and transfer cost to around $55 – a reduction of over 80%. There is also a noticeable reduction in the overall transfer time, from 39 to 34 minutes, meaning potential savings in uptime.
Table 1. Summary of simulation results.
Cases 2 and 3 investigate waterless transfer – Product B is used directly to displace Product A. Case 2 summarizes a conservative approach that re-jects all material that is not close to specification. With no extra water pre-sent, the amount sent to drain is much reduced (1175kg vs. 2810kg), so the concentration is increased – typically beneficial if an anaerobic digester is available. Since the cost of water and waste disposal is relatively low, though, the overall savings for this approach are also relatively modest – around $10 or 3%. Case 3 recovers more of the transition material into the two product tanks, again maintaining their compositions within 1% of the specification. As with Case1, significant overall savings are achieved (84%); the volume of waste produced is less than half that of Case 1; and the transi-tion time is reduced 15%.
The lower part of Table 1 shows the impact of the transition cost on the sim-ple raw material cost of Product 1 (15.8¢/kg). As can be seen in the Base Case column, adding the transfer costs to a batch of 25 tons increases the apparent cost of production by 8% to 17.1¢/kg. If the batch size is doubled, the unit cost can be reduced to 16.4¢/kg, while accepting a 1% deviation from the specification can almost eliminate the overall transfer cost impact.
While this analysis investigates a simple process and only one product tran-sition, it demonstrates an approach to understanding the impact of different approaches to product transition. It can quantify the cost, waste and uptime implications of those strategies and provide improved understanding of the cost implications of run length. In each of the real-world facilities where we have applied this type of analysis recently, CPO has achieved capital-free savings of over $200,000/y, as well as providing capacity enhancements.
When to consider
This approach is particularly well suited to analyzing the cost and implications of transitions between different products running through common equipment. It can also be very useful in un-derstanding the impacts of disturbances or step-changes in large continuous processes.
About the Author
Charles Sanderson is the Technical Director of CPO, based in Minnesota, USA. He has led the development of Cargill's simulation capa-bilities for over ten years. He and the team have deployed the tools in projects across Europe, Asia, Africa and the Americas. Charles graduated in Chemical Engineering from Imperial College, London and received his PhD from Sydney University in Australia.
With more than 500 projects completed that have less than one year to payback, Cargill Optimizing Services understands how to decrease costs and increase yields in the food and beverage industry.