Analyze the trade-offs between multiple supply chain objectives for faster, more informed decision making
The Challenge of Competing Business Objectives
Certainly minimizing cost is a common business objective, but in the real world there are many other objectives driving supply chain design which are often in conflict with each other. In addition to maximizing profitability, companies are also challenged with objectives such as best-in-class service, risk robustness and production flexibility, among many others.
To add to the complexity, the trade-off between two competing objectives is not always easily predicted. To identify the trade-off users often create and run a multitude of what-if scenarios to find the optimal choice to satisfy both objectives. Supply chain analysts are look- ing for high-level answers about the trade-offs between solutions prior to creating detailed what-if scenarios.
The Solution: LLamasoft Multi-Objective Optimizer
The Multi-Objective Optimizer (MOO) is an integrated tool within LLamasoft® Supply Chain Guru® that allows customers to meet business challenges that often include multiple conflicting goals. Using a simple user interface, the tool automates the analysis of trade-offs across 19 available objectives. Multi-Objective Optimizer automatically optimizes to and provides insight into any two best solutions (i.e. lowest cost and best service) then quickly presents the tradeoff analysis between the two.
Useful in any supply chain decision with more than a single objective, the tool builds a trade-off curve allowing analysts to visualize the interaction and span between different options and more efficiently move to detailed scenario analysis.
Select the best points on the curve for full scenario analysis
LLamasoft Offers a Simple Workflow for Optimizing for Multiple Objectives
- User selects first and second objectives
- MOO automatically creates Pareto curve
- User selects the best points on the curve for full scenario analysis
Use Case examples:
Leveraging the Power of Multi-Objective Optimization
Example 1: CO2 Emissions vs.Total Cost
When evaluating the impact to overall cost as companies consider reducing carbon footprint, it is often difficult to understand the relationship between cost and emission levels.
By applying multi-objective optimization, one can clearly see what level makes the most sense to target (see graph). By reoptimizing the existing supply chain, a reduction of roughly 5M tons of CO2 can be achieved at almost no cost. Furthermore, by making a limited investment of $1.8M (roughly two percent of total cost), this company could reduce carbon output by an additional 10M tons or equivalent to a roughly 30 percent reduction in emissions. While further reductions are possible, the cost trade-off is much less attractive, as seen on the left side of the graph.
Example 2: Number of Transportation Assets vs. Weighted Average Transportation Lead time
To provide the best in class service, how many trucks should I keep in my fleet? What’s the impact to service when I add the next dedicated asset?
These are questions often asked by businesses in order to understand the impact of increasing or decreasing the number of transportation assets. Rather than focusing objectives purely on cost or profit, the example graph above left illustrates the impact on service when vehicle counts between 84 and 89 are evaluated.
The output shows that the impact of adding vehicles is most dramatic between 84 and 87, where an average reduction of 2.5 percent or 2.3 billion unit miles can be achieved per vehicle increase. However, the improvement at the last incremental step between 88 to 89 vehicles is much less dramatic at .09 percent or 82 million unit miles. Therefore, to justify the spending on the last two vehicles, one may need to consider other business and financial objectives such as additional capital outlay, service offered by the competition, variable cost/savings related to incremental vehicles and associated capacities, to name a few. From that perspective, multiple MOO graphs can be generated to support the analysis—such as the Number of Assets vs. Total Supply Chain Costs graph above right.