Lawrence W. Robinson, Professor of Operations Management at Cornell University’s Johnson Graduate School of Management took Nate Silver’s US Senate prediction one step further by adding Monte Carlo simulation to the mix.
In recognition that the decisions it needs to make around business-critical innovation are highly complex, global fast moving consumer goods supplier Unilever developed its Decision Making Under Uncertainty (DMUU) approach. Combining a structured method with Palisade’s DecisionTools Suite software ensures that project teams fully understand the scope of their decisions, and have the tools and the knowledge to make informed and high-quality choices. This prevents opportunities and threats being overlooked, and increases Unilever’s agility in the market place.
Unilever is one of the world’s largest suppliers of fast moving consumer goods in the refreshment, foods, home and personal care sectors. With a portfolio of over 400 brands, it has consistently ambitious growth targets. The company has an extensive annual budget for cutting-edge research and development, and thousands of projects in its innovation pipeline at any one time. This means that in order to make informed decisions on how to manage this portfolio, it needs absolute clarity around the risks and opportunities it faces.
However, like any large, dynamic organisation, complexity has a large impact on Unilever’s decision-making process. Many parties are involved in the process, often with conflicting values, motivations, perspectives, personalities and power bases. These organisational complexities are reinforced with analytical complexities such as the large number of interrelated inputs that must be factored in to the decision, the high level of uncertainty inherent in early-stage developments and potentially conflicting decision criteria.
A Structured Approach to Decision Making
For business-critical innovation, Unilever recognised the inherent complexity of its decisions and the need to maintain a dual internal and external focus to prevent important opportunities and threats from being overlooked. It understood that incorporating these factors into an effective decision making process would improve decision quality, facilitate faster decision making and ultimately increase Unilever’s agility in the market place.
The Unilever response was to develop a unique approach known as Decision Making Under Uncertainty (DMUU). This is a disciplined, methodical and structured approach to decision-making, with probabilistic analysis at the heart of its logical reasoning. It combines framing and structuring tools with leading-edge analytical software – Palisade’s DecisionTools Suite. The DecisionTools Suite is an integrated package of seven risk, decision, and data analysis tools that run in Microsoft Excel. This approach ensures that project teams fully understand the scope of the decision, that they have the tools and the knowledge to make high-quality decisions, and the insight to understand the consequences of taking one course of action over another.
Overall, DMUU helps to provide the required clarity, insights and commitment to action.
DecisionTools Suite Guides Decisions on Innovation
Unilever selected Palisade’s DecisionTools Suite as the principle analysis software to support its DMUU process and decision-focused culture due to its flexibility and ability to do Monte Carlo and decision tree analysis using component products @RISK and PrecisionTree, respectively. Today, the DecisionTools Suite enables Unilever to develop probabilistic business cases for its biggest innovations, as well as its major strategic decisions.
DMUU and the use of the DecisionTools Suite is now a standard part of Unilever’s innovation process and probabilistic business cases are required for all big and complex projects. For example, a typical use for @RISK, the risk analysis element of the suite, is in evaluating alternative strategies for a new product launch or a major capital investment.
Unilever teams also use PrecisionTree, the decision analysis tool, to evaluate early stage projects where decisions and uncertainties will occur at various times in the future. This approach, using decision trees in PrecisionTree, is used to evaluate the current value of a project and also to understand the risks and benefits of internal versus external development routes.
In recognition of the importance of the DMUU, Unilever has an internal consultancy function to provide decision support and software expertise when required.
In addition, Palisade’s software is used to support other business areas including supply chain, safety, regulatory, as well as additional complex one-off decisions. All of these have the common features of multiple compelling alternatives, significant contradictions on how to proceed and high stakes should the ‘wrong’ decision be made.
“Strategic decisions require a process that addresses all the elements of decision quality,” explains Andrew Evans, decision analyst at Unilever. “However, an integral part of that process is powerful and flexible software that informs the debate on which direction should be taken. We evaluated various options and Palisade’s DecisionTools Suite was the tool that best met our business requirements. As a result it has played a key role in increasing the quality of decision-making and helping project teams to think clearly, act decisively and feel confident.”
An Invaluable Asset for Student Field Placement Management
Decision Analyst, Unilever
Key software / features useful to Unilever: @RISK is the most commonly used application of the tools available in the DecisionTools Suite. Decision-makers at Unilever are now used to seeing insights from business cases described using histograms and advanced sensitivity tornados. Box-and-whisker diagrams (box plots) are also very useful when alternatives or projects need to be compared. Sensitivity and scenario analysis are used to understand the key drivers of uncertainty. In addition, analysts help to draw insights from the models using summary graphs and scenario analyses.
Pert and Triang are the distributions used most often when Unilever is deploying @RISK to evaluate business cases, as they are good for describing distributions when data is elicited from experts. The discrete distribution is used to simulate alternative futures, such as competitor action, or different levels of success in a product launch. However, when good quality historic data is available, or when the ‘tails’ (eg in safety studies) are of interest, Unilever uses the wider set of distributions and tools such as distribution fitting feature available within @RISK.