Copperleaf provides decision analytics to companies managing critical infrastructure. Our optimisation capabilities, as described in my previous post, empower our clients to make the highest value investment decisions. But we often get asked ‘can you quantify the benefits of optimisation?’ or ‘how much better is optimisation than the prioritisation process I currently run in Excel?’.
To answer this question, we can certainly point to our existing client base. For example, one organisation obtained $42 million of additional value in their portfolio through our optimisation techniques. However, there is an obvious reluctance to rely on anecdotal evidence from a handful of peer organisations with their specific operating context.
With this in mind, Copperleaf sponsored a MSc. research project led by Ibrahim Tamimi and supervised by Dr. Patrick Beullens with the Centre for Operational Research, Management Science and Information Science of the University of Southampton in the UK. The objective was to quantify the benefits of investment portfolio optimisation versus prioritisation for asset-intensive organisations.
The researcher first constructed a methodology to generate representative portfolios of investments. An investment is basically any project or intervention activity carried out to bring value to the organisation. Actual investment data provided by various asset-intensive organisations was used to ensure these portfolios were as representative as possible of candidate investment portfolios in these types of organisations. Two scenarios were then tested:
- A traditional prioritisation algorithm
- A Mixed Integer Linear Programming (MILP) optimisation
Repeating this many times, the results are effectively a Monte Carlo simulation of both investment portfolio prioritisation and optimisation which can be analysed to understand if there are difference in overall value between the two techniques.
We also completed a factorial analysis and considered a number of factors we thought were significant:
- Number of investments within the portfolio
- Allowing investments to have multiple alternatives. Alternatives are different investment options to mitigate the same risk or meet the same need (e.g. refurbish, replace, change in maintenance regime etc.)
- Duration of the constrained period (e.g. optimising to produce next year’s investment plan or a 5-year regulatory business plan)
The results show that optimisation always yields higher portfolio value for the same monetary constraints compared to the prioritisation model. This is consistently in the range of 7% to 20%. Secondly, the advantage of optimisation intensifies as the portfolio complexity increases. For example, the presence of multiple project alternatives roughly doubles the optimisation benefit, and a multi-year constraint increases the optimisation benefit by about 50%.
7% to 20% of increased value is a significant amount—especially given the size of typical investment plans in asset-intensive industries. Hopefully this gets you thinking about whether investment portfolio optimisation could be relevant for your organisation. Or, if you are already convinced, this generalised result could be used to support business case development for a wider asset management transformation.
Want to find out more?
Ibrahim Tamimi and I presented this research paper at the IET Asset Management Conference in London. Feel free to download a copy of the published paper.
In the meantime, please do get in touch if you would like to find out more. Also, if you have your own examples of the benefits of optimisation or asset investment planning, I would love to hear from you.