Not all models are the same
The quality of wastewater hydraulic models can vary from very general models with few details to meticulous and site-specific models which can be used to refine and challenge different scenarios.
Understanding the performance of wastewater infrastructure, seeing where the problems are and why, is highly complex and involves being aware of a large number of variables. Making the correct judgements about these variables is fundamental to how accurate the model is and ultimately how useful it is.
The most common mistakes modellers can make are:
- Inaccurate system attributes
- Misunderstanding objectives and budget requirements
- Economising on monitors
- Failing to optimise accuracy
- Inadequate experience
1. Inaccurate System Attributes
If input data is missing or incorrect, this can have a significant impact on the way the model replicates real world situations. The complexity of the infrastructure is such that having a full and complete set of highly accurate physical attribute data for the entire system is frequently not possible. Even when data is available it may not be in a common format, making the process of importing it into a model complex and time consuming. Consequently it is easy to consider the impact on the model if the wrong information is fed in, in terms of incorrect pipe diameter, lengths or connections.
2. Misunderstanding Objectives and Budget Requirements
Modelling can be carried out at varying levels of complexity and the level of work involved will necessarily reflect in the cost of the modelling work. It is therefore of vital importance to understand the clients objectives or desired outcome and align these with the budgetary limitations. For example offering or indeed setting the parameters for an unsustainably low tender-price can lead to serious problems in the development and application of wastewater models.
Of course competitive pricing is important, but a price only approach to tendering wastewater modelling contracts is not to be recommended. Lowest price is not always best value and time and again it is the case that it results in greater cost to the client. This is because not only do they have to spend more money to get a model of sufficient quality to gather the information they need, (sometimes paying to have models re done) but also a poor model may result in expenditure on the wrong capital solutions. Capital may be spent on infrastructure which does not perform due to inaccuracies or unobserved results in the original model.
3. Economising on Monitors
There is a very simple rule when it comes to flow monitors and that is if you try to save money and effort in this exercise it will be detrimental to the modelling work; a case of getting what you pay for. Spending less will result in a lower quality of data which ultimately results in a model which will be less reliable and will give less confidence to its predictions and ultimately to any decisions which are based on the model outcomes. It is essential that time and therefore money is spent on ensuring the flow monitors are in the right locations, that they record accurate data and that the data is sufficient in quantity on which to base decisions.
4. Optimising Accuracy
In some ways this relates back to number 2 and whilst there would always be a desire for a model to be accurate the level of detail required may not always be to the highest. The modeller should be looking to deliver a model which is as accurate as it needs to be. Consequently it is important to be clear what the objective of the modelling work is and what questions need to be answered by it, before the modelling work begins. It pays in this case to think longer term to recognise what questions or issues the model might be required to address further down the line. As the model is generally built from the bottom up knowing what is required from it, will help define the appropriate level of accuracy.
5. Modelling Experience
As outlined above the modelling process incorporates a large number of variables so perhaps the most significant area of error on wastewater modelling projects is due to inadequate modelling experience. Problems can arise from inexperience both when this arises on the client side as well as from the consultant. Whilst it is undoubtedly more obvious that the modellers themselves should be appropriately experienced there will also be shortcomings if the client does not fully understand the analysis tool at hand. It is important that experience should include time in the field and not just use of modelling software to fully appreciate the implications of changes in the variable applied. An understanding of the meaning behind making changes in the model and how to tell if they are right or wrong improves with experience. If the client understands the modelling software they are better placed to interpret results or challenge assumptions. A good modeller will be able to articulate for the client the issues and assumptions associated with any particular model.
Our modellers specialize in integrating GIS with models using InfoWorks ICM which offers exceptional results through its general versatility and functionality. It provides us with the ability to test future developments through the use of ‘what if flooding scenarios’. We are also able to work closely with engineers, operators and designers to ensure that the gap between models and reality is minimised.