A greying fellow reflects on... Sustainability
Part 2: Life cycle costing and carbon modelling
George Bothamley FCInstCES
ONE of the members’ pledges in the CICES sustainability policy is to adopt life cycle costing, whole-life carbon modelling and post-construction modelling. These are concepts that I worked with on capital water industry projects from the early 21st century, and this article will reflect on some of these experiences with a view to providing some insights that you may develop in your achievement of this pledge.
I first encountered a contractual obligation for life cycle costing on a capital water project framework in the early 2000s.
The water company sought to improve capital efficiency by setting whole life target costs for capital schemes, based on what similar schemes would have cost historically, but with an efficiency factor applied. The contract partner was challenged with finding an even cheaper scheme to enable it to achieve gain share. The capital cost was calculated using a database of historic component costs. The lifetime cost component was calculated fairly crudely, using average costs as a percentage of capital cost for mechanical and electrical components and capital replacement costs for civil, mechanical and electrical costs for life-expired components. Thus, the contract partner was incentivised to come up with increasingly efficient schemes.
Efficient schemes
My first involvement was to attend solutions option meetings where a selection of possible designs for a project were discussed and one selected for taking forward. No costings were presented at these meetings; merely designers stating what they thought were the best and cheapest solution and a consensus on the scheme to take forward emerging.
Life cycle costing was only done for the chosen scheme and compared with the historic scheme upon which the target was based. I was shocked at this. I had been used since the 1970s and 1980s to be presenting comparative costs for alternative solutions to elements of design and construct civil engineering schemes. For the capital water projects a simple spreadsheet to provide comparative costs was quickly introduced. Five or six potential solutions could be costed in two or three hours before the meeting and better decisions consequently made.
On a later capital framework, someone with much better IT skills than mine developed a more sophisticated spreadsheet-based calculator that was just as quick to use. It dealt with operating costs more accurately and also dealt with carbon effect which was then becoming a consideration. This worked very well, almost certainly contributing to better option selection and enabling more data analysis.
On a subsequent framework, electronic wizards came up with a complex tool that delivered a very detailed assessment of whole life costing and carbon effect. But few people had sufficient skill to implement it, it took relatively long to produce reports, and was difficult for a lot of people to follow at meetings. So it was only used for the scheme option selected, if at all. In my view our sophistication had managed to take us backwards ten or twelve years.
We need a tool that is quick and simple so that it is extensively used.
Although challenging to retain simplicity despite the addition of more data, it should be possible to produce such a tool, and to make it much better than those we used in the pre-digital age. Later we will reflect on how you might contribute to this. But first we will reflect on some observations from calculating life cycle costing and whole life carbon calculations undertaken on water capital frameworks earlier this century which illustrate why we need such a tool.
Lifetime costing
Without undertaking lifetime costing, confirmation bias can easily occur. For instance, an expectation frequently expressed by managers is that life cycle costing would identify schemes where a higher capital cost will be more than offset by reduced operating and capital replacement costs. Employers love this concept; they all seek to reduce both capital expenditure and on-going maintenance costs. In practice this was found rarely to be the case.
Without undertaking lifetime costing, confirmation bias can easily occur.A decade or so ago I reported on a practical example of this erroneous thinking. A design assumption on small sewage works was that progressive cavity pumps had in-use savings that more than justified the additional capital costs compared to simple fixed-flow rate pumps.
To overcome an issue with low flows that progressive cavity pumps could not solve, a small works was designed with fixed speed pumps which fed forward at times excessive flow, with the excess returned and pumped again. Cost comparison showed that this solution not only reduced first capital cost, but also the in-use costs were less energy and maintenance costs were, perhaps surprisingly, lower, as were periodic replacement costs. Fixed speed pumps were subsequently used on many schemes on the framework with considerable lifetime cost savings.
Another example was on a large raw water intake pumping station. The life-expired pumps being variable speed were assumed to be more efficient than fixed speed pumps to allow for different allowed abstraction rates according to river levels. They were configured as duty, assist and standby. A solution using much cheaper fixed-speed pumps in a duty, assist, assist/standby configuration was found to be capable of covering the different allowed abstraction rates with reasonable efficiency and lower operating and maintenance costs.
These examples had both lower capital cost and lower in use costs. Many other schemes on the framework selected solutions with similar life cycle cost profiles to the above examples.
A robust and roughly right calculation of life cycle costs could be quickly made for all options being considered for a scheme
A report I made on operating cost modelling illustrated these and other examples such as a technically attractive scheme not providing best value (electro chlorination compared to bulk hypochlorite dosing providing operating cost savings that were less than the additional capital cost) and optimising pipeline size (where a larger main reduced pumping costs but not by as much as the additional capital cost). But solutions with increased capital cost being more than offset by operating and maintenance savings were extremely rare.
These examples had both lower capital cost and lower in use costs. Many other schemes on the framework selected solutions with similar life cycle cost profiles to the above examples. A report I made on operating cost modelling illustrated these and other examples such as a technically attractive scheme not providing best value (electro chlorination compared to bulk hypochlorite dosing providing operating expenses savings that were less than the additional capital cost) and optimising pipeline size (where a larger main reduced pumping costs but not by as much as the additional capital cost). But solutions with increased capital cost being more than offset by operating and maintenance savings were extremely rare.
A false assumption commonly made when life cycle costing comparisons are not carried out is that the solution with least capital cost is the most economic. A cost consultant to a water utility was obsessed with preventing contract partners getting unearned gain share, and they considered that the contract partners often did nothing to achieve operating cost savings. The commercial model for gain-share was amended to be largely capital expenditure only, resulting sometimes in the selection of low capital cost schemes with higher whole life cost.
In our pre-digital life-cycle costing initial capital cost and life-expired replacements were assessed for eight or ten component categories. For operating costs, standardised methods of assessment for sludge density, chemical prices and dosing rates, energy cost and usage, annual maintenance of M&E equipment, statutory rates and charges, and staff enabled quick and roughly right calculation.
The metric for all items was £ sterling. Later, a more automated spreadsheet was created and carbon effect added. Thus, a robust and roughly right calculation of life cycle costs could be quickly made for all options being considered for a scheme, enabling an informed decision on the most sustainable solution to be made and monitored.
What next?
To help with this in future, it would be good to have an updated tool fit for the digital age. Although the more complex data including more sustainability and environmental considerations will make this difficult, it should be possible for us to produce a simple and effective app. A future reflection we will look at ways of achieving this, but in the meantime, we will reflect on some other aspects of sustainability.
George Bothamley FCInstCES