Geotechnical processes are the key to innovation in civil engineering but without full understanding their potential is limited, argues Robert Mair.
Civil engineering often involves geotechnical processes but in many cases practice is ahead of our understanding of the fundamentals of the process. Here are a few examples:
Grouting - where does the grout actually go to? How can we predict whether or not the ground will fracture in the process of grout injection, and in which direction will the grout travel? How does the stress history of the ground influence the behaviour of grout injected into it? Is compensation grouting better suited to some types of ground than others, and if so, why?
Soil conditioning additives and lubricants - how do these work? What additives are suitable to condition particular soils to enable earth pressure balance tunnelling machines to operate successfully? What are the mechanics of pressurised spoil being transported through the screw conveyor of a tunnelling machine? How do 'lubricants' injected to reduce pipejacking forces really work?
Forepoling, umbrella arches and face nailing in tunnelling - how do these techniques really work? Are there reliable ways of calculating their influence on tunnel face stability? What influence, if any, do they have on reducing ground movements caused by tunnelling?
Micropiling - how do groups of small diameter piles interact to form effective foundations or retaining structures? Can the diameter, spacing and inclination be optimised? Can meaningful calculations be undertaken to quantify reliably the forces acting on micropiles?
In all these examples, the application of the particular geotechnical process is largely empirical. In some countries, particularly Italy, France and Japan, there have been notable cases of major innovation using geotechnical processes - even though understanding of the mechanics of the process may be far from clear.
Many successful projects have been achieved through the introduction of novel geotechnical processes without there being full understanding. The use of umbrella arches in tunnelling and micropiles for foundations and retaining structures are just two examples.
Does it matter if the detailed mechanics of the process are not all that well understood?
Provided it works, who needs to know the exact details? After all, there have been many inventions and innovations throughout history that have not been fully understood but have more than fulfilled their purpose.
The problem with not fully understanding a process in today's world is the limit this places on innovation.
If, as a profession, geotechnical engineers wish to introduce innovative techniques to make a big difference to projects, more than ever before we have to convince ourselves and others that these are safe and robust. To do this, we need to be able to understand the fundamental behaviour and mechanics of a geotechnical process, so that meaningful design calculations can be done - in many cases to quantify properly the safety of the process.
Without convincing design calculations, sceptical clients are entitled to remain sceptical. It is no longer acceptable for us to use particular geotechnical processes saying 'we don't really understand how this works, but we think that it will probably be OK'. We have to do better than this if we want to take innovation forward.
There is no substitute for field measurements to assist our understanding.
But all too often these are in short supply, and those we have are inevitably limited to a few particular sets of ground conditions and geometries. Importantly, w e do not normally have measurements of processes where things are going wrong - in other words when failures are imminent.
We should use modelling to really understand the detailed behaviour of a geotechnical process, when it is working well and when failure is happening.
Numerical modelling offers enormous opportunities to explore a wide range of parameters, and complex 3D situations are increasingly within our reach to analyse reliably (although in the wrong hands sophisticated computational analyses can be very dangerous).
Physical modelling, either at enhanced gravity levels in a centrifuge or on the laboratory floor, allows detailed examination of geotechnical processes. It also provides a considerable amount of physical measurements against which numerical modelling can be tested.
Ideally, we need the combination of all three: field measurements, numerical modelling and physical modelling. We often cannot get all three, but we need at least two of them to make real advances in understanding of the mechanics of geotechnical processes. Without this we will be limited in the novel techniques that can be safely introduced. It is only with this understanding that we will make the most of the opportunities to innovate.
Robert Mair is professor of geotechnical engineering at Cambridge University and head of the civil and environmental engineering division. He is a founding director of Geotechnical Consulting Group.