How Safe Is Your Floor? [Jul 2006]
Slipping on floors results in many injuries - the cost to industry in terms of the disruption, worker time off and compensation is thought to be huge - the UK's Health and Safety Executive has estimated it at £512 million.
Most people imagine that the science involved in slipping is fairly mundane and indeed in some respects it is. In other respects it is complex, so complex in fact that very few people really understand it. The problem is that it involves friction and not just normal dry friction but partially lubricated friction. Unfortunately this lack of understanding has led to something of a muddle, with misconceptions being the norm rather than the exception.
First, a little bit of history
Back in the late 1930s, early 1940s, Percy Sigler in the USA was given the task of measuring the slipperiness of many of the floors in government buildings in Washington. He designed a machine called the Sigler Pendulum. Having used the device and taken many hundreds of readings he found that generally floors which had a dynamic coefficient of friction below 0.4 had a history of accidents, whereas those with a coefficient higher than 0.4 were almost invariably accident free.
In the 1950s, Barbara Sabey of the UK Transport and Road Research Laboratory visited the USA, saw Sigler's machine and realised it could be used to measure the skid resistance of road surfaces. She brought it back to the UK, redesigned it, and as a result that instrument, known as the TRL Pendulum, is still in use today measuring road surface skid resistance throughout the world.
In the early 1960s the Greater London Council who used the Pendulum for road surface investigation decided to use it for measuring floors not realising its original use. Over the next few years they measured some 3,500 floors throughout London and found that floors on which slipping accidents had occurred gave readings less than 40 (dynamic coefficient 0.4) whereas if the floor gave a reading higher than 40 it was almost accident free.
Coincidentally, the UK Building Research Establishment were carrying out work on simulating wear on flooring materials. They decided that they ought also to find out what forces people actually exerted on the ground or floor as they walked. It was this latter piece of work which became far more important than the work on wear.
In carrying out their tests, the BRE arranged for some 60 members of their staff to walk along a specially constructed catwalk and in the process to step onto a force plate which measured both vertical and horizontal forces. While the force plate was crude by today's standards it nevertheless gave very informative data.
It was found that people need a maximum level of friction to be developed at two points in their stride. The first is when the heel of the leading foot which is swinging forward first meets the ground and the second is when the sole of the rear foot is pushing the body forwards just before it leaves the ground and swings forward. Whilst this may sound obvious, no one until then had actually measured it.
They were aware that in slipping the most dangerous moment is the heel strike. If the foot does not find sufficient restraint it slides forwards and as a result requires more and more friction to be able to restrain it. Because the frictional capability of the floor/shoe interface has already been surpassed causing the initial slip, the slip becomes uncontrollable and the person falls to the ground.
The situation with sole slip is much more controllable and indeed to most people it is merely an annoyance when it occurs. It quite often happens when pushing a heavy trolley such as in a supermarket - it very rarely leads to an accident.
By analysing the forces involved in the heel slip situation, the BRE found that of their 60 volunteers 50% required less than 0.19 coefficient of dynamic friction and 50% more than 0.19. The minimum value recorded was around 0.08 while the maximum was 0.28. Statisticians then analysed all this data with the aim of predicting what would be the maximum value they would have recorded had they used 1 million subjects rather than only 60. The result was a figure of 0.36.
Whilst carrying out the simple walking tests they also carried out tests to see what happened when people turned during the walking process. By similarly analysing the results they found that 1 in 1 million people would require 0.38 for the dynamic coefficient of friction. The value of 0.4 was therefore adopted as representing the criteria for safe walking. In other words, if the floor/shoe combination gave less than 0.4 it was a potentially slippery situation, whereas above 0.4 it was 'safe'.
As will be appreciated this fitted in almost exactly with the results of the work of the GLC and, indeed, Sigler many years previously. Thus, from the 1970s the Pendulum started to be used more frequently for floor slip accident investigations using this criteria.
Introduction of the Tortus
In the 1980s the Ceramic Tile Research Laboratories in the UK decided to produce a more sophisticated machine to measure slip resistance than the rather clumsy difficult-to-operate Pendulum. The result was the Tortus. The Tortus was an excellent machine from a technological point of view. It appeared to do all the right things - it measured the friction developed by the slider in the conventional way as it trundled over the floor. The only problem was that the results it produced in the wet did not correlate at all with the Pendulum. This was not a new phenomenon. Several investigators at that time relied upon pulling a weighted shoe across the floor and calculating the coefficient of friction from the force needed. They too had found little or no correlation with the Pendulum. The question was which instrument or method was correct.
This author, and at the same time the Research and Laboratory Services Division of the Health & Safety Executive, carried out some research to try and find the answer. We both concluded that the answer lay in the physics of lubrication. We realised that the test machine must react to the upthrust produced by the film of water on a surface when a body moves over it in exactly the same way as under the heel of a slipping pedestrian.
HSEL then calculated a unifying factor called the 'critical film thickness' for the Tortus, the Pendulum and a slipping pedestrian and pronounced that both the Pendulum and the Tortus were wrong. They suggested that the Pendulum's critical film thickness was too large, while that for the Tortus was too small compared with the slipping pedestrian. This author spotted that some of the measurements used by HSEL for the Pendulum were incorrect and that when the correct figures were used, the resulting critical film thickness produced by the Pendulum matched that of a slipping pedestrian. Indeed, if this had not been so it would be most unlikely that the correlation between Pendulum readings and the accident record of floors could be as good as it was known to be.
Three or four years ago the author decided to take the matter one step further. If the hydrodynamic film theory was correct it should be possible to design a machine specifically to correlate with the Pendulum but use different principles of operation. The result was SlipAlert, and confirmation that the hydrodynamic film theory is indeed correct. The advantage that SlipAlert has over the Pendulum is that it is much easier to operate, quicker and significantly cheaper.
Whereas previously only the largest industrial organisations who had their own testing facilities would themselves own a Pendulum and occasionally test their floors, now with SlipAlert even a small company can do so on a regular basis.
Why not use a Pendulum?
Although it is a fine instrument it does require an experienced operator to use it. That operator needs to follow a strict protocol as laid down in BS 7976 Part 2 or the UK Slip Resistance Group Guidelines. The machine itself costs around £3,500 and £300 per annum to have calibrated. To set it up and take a set of readings will probably take around 10 to 15 minutes per location. If you do not have your own Pendulum you can hire one but you do need to know exactly how to operate it in order to be confident of your results. The alternative is to get a Test House to come and take the readings for you. In the case of accident investigation it is essential for an independent body to take such readings, but because the service of a Test House is likely to cost several hundred pounds it is not a practical or realistic method for regularly monitoring a floor or a number of floors.
SlipAlert overcomes the problem by allowing the Health & Safety staff member to do the testing on a regular basis at a reasonably cost.
There are other machines available but in the experience of the author and bodies such as the UK's Health & Safety Laboratory (HSL) they do not give credible answers in wet conditions. HSL have attempted to use roughness measurement combined with a computer system (SAT - Slips Assessment Tool) as an alternative. The problem is that roughness is such a complex parameter and as yet, although one can see in certain situations some form of correlation with wet slip resistance, it cannot be relied upon. The UK Slip Resistance Group in its latest Guidelines warns 'It is imperative that roughness measurements should not be relied upon themselves to judge the likely slip resistance of the floor. Some reliable means of directly measuring the slip resistance in wet conditions should always be used in conjunction with roughness measurements'. In respect of SAT, HSL wrote in the UK Slip Resistance Group Guidelines 'HSE clearly states that it should not be used as the basis of floor specification or modification, nor should it be used in legal proceedings'.
Whilst SAT of itself is a very useful tool, we have been advised by HSL that they are intending to link it with direct readings of slip resistance from instruments such as the Pendulum and SlipAlert. SAT, together with the Pendulum or SlipAlert, will make it an extremely powerful tool.
This leads on to the question of how one decides whether a floor is satisfactory or safe. In the majority of cases if a floor has a slip the coefficient of friction from the force needed. They too had found little or no correlation with the Pendulum.
Whilst SAT of itself is a very useful tool, we have been advised by HSL that they are intending to link it with direct readings of slip resistance from instruments such as the Pendulum and SlipAlert. SAT, together with the Pendulum or SlipAlert, will make it an extremely powerful tool.
This leads on to the question of how one decides whether a floor is satisfactory or safe. In the majority of cases if a floor has a slip resistance in the wet and dry of 0.4 or greater when tested with the Pendulum or SlipAlert fitted with a Four S slider, then the potential for it to cause a slip can be regarded as 'low'. In practical terms most people would regard it as 'safe'.
However, many floors simply do not achieve 0.4 in the wet and whilst in simplistic terms these might be considered less than 'safe', in practical terms they can be accident free and acceptable for use in that situation. Indeed, many thousands of floors in the UK could be considered less than 'safe' because they fail to reach 0.4 in the wet - that does not mean that we need to replace them or regard them as unacceptable.
It should also be understood that there are situations where 0.4 is too low a criterion for safety. Although these represent only a small percentage of common situations they also need to be recognised. How does one therefore go about deciding what is the criterion to use for a particular floor?
The answer to the following twelve questions will enable an assessment to be made which will give a good indication as to whether a floor could be regarded as acceptable even though it may not achieve 0.4 in wet conditions. For convenience the questions are arranged in four groups.
1 LOCATION
1.1 - What is the floor used for? (ie. what activities take place)
- If it is a 'normal' floor where people walk around, turn and walk at normal speeds, then no modification is needed
- If it is a narrow passage or corridor where people simply walk, generally without turning, then 0.36 is an acceptable criterion
- If it is an 'activity' floor where people might be running around, then one may need 0.5 or even higher, depending on the particular activity
- If it is a 'dance' floor one may need to ensure that the slip resistance is not only not too low (usually around 0.36) but not too high (ie. not more than about 0.45)
1.2 - Is the floor level or on a slope?
- If the floor is on a slope people need more friction when they walk down the slope. One needs to increase the criterion on slopes by Tan where is the slope angle
1.3 - Is the floor such that in all lighting conditions one can see liquid contamination on the surface and the floor is never sufficiently crowded that a person has a restricted view of the floor or there are other distractions around such that a person might not see the contamination?
- People can change their gait and require significantly less friction if they believe the floor is likely to be slippery. If the contamination is only occasional, eg. a spillage, then people can either avoid it or walk defensively. This question needs to be considered in relation to 3.2.
2 USERS
2.1 - Who is likely to use the floor?
- The original 0.4 criterion was based on fit, active people. If disabled people use the floor one needs to be aware that they may need frictional restraint at the higher end of the scale. Similarly elderly people may need similar restraint to a younger person, but their reaction to a mini slip will be much slower and unlike a younger person they may be unable to correct it before it develops into a full blown slip
2.2 - How many people use the floor?
- On a purely statistical basis, if only a few people use the floor then the chances of any of them being that 1 in 1 million person who requires 0.4 is small. Indeed, the maximum that any of the 60 subjects in the BRE test required was 0.28. In a domestic situation or similar it is not unreasonable to reduce the criterion down to 0.3, particularly if all those using the floor are very familiar with it and, for instance, know not to walk over it when it is wet.
2.3 - What footwear is likely to be used?
- The slip resistance value is based on a normal good quality rubber type heel. However, some relatively common heel materials can give very different values of slip resistance. For instance, some heels found on ladies' shoes can give very low values in both wet and dry situations. Leather is generally satisfactory but often only just so in the dry compared with rubber. Soft rubbers such as found on trainers can give lower slip resistance in the wet although better in the dry than the harder rubber normally used to measure slip resistance
- If the floor is open to the public you may well need to consider how it behaves against these poorer forms of footwear
- If it is a changing room or swimming pool then people will be barefoot
- If access to the floor is restricted to a few employees, a very worthwhile means of overcoming a potential slip problem is by the issue of special shoes/boots. One particular make has been found to be effective in this respect. However, it must be appreciated that most 'safety' shoes/boots are no better (nor worse) in respect of their slip characteristics than normal shoes. The 'safety' aspect relates to protecting the foot rather than the person
3 CONTAMINANTS
3.1 - What contaminant is likely to be involved?
- By far the most common contaminant is water. However, virtually any free flowing liquid can act as a lubricant in slipping. In a supermarket, for instance, many types of contaminant can be spilled onto the floor by customers accidentally dropping goods onto the floor. Powders such as flour, sugar, dust etc also act as lubricants
3.2 - How frequently does the floor become contaminated?
- Some floors are frequently wet, for example swimming pool surrounds, changing rooms, etc. Others get wet on a fairly random basis, eg. floors with direct access from outside, toilet areas where users drip water from their hands onto the floor. If the frequency is very low and perhaps only occurs during cleaning, it may be appropriate to consider the floor 'dry'
3.3 - Are measures taken to prevent contamination?
- Measures such as properly designed door matting can make a significant
difference to the frequency and amount of contamination reaching the main floor. They do however need to be properly designed to be really effective. Remember that sufficient water can be brought in on an umbrella which the user shakes onto the floor once he has walked over the mat, in order to cause the next person to slip over!
4 CLEANING
4.1 - Is the cleaning not only appropriate but likely to be effective?
- Many floors, in particular floors based on synthetic resins, need mechanical means to clean them; in other words, not a mop and bucket! Lack of proper cleaning can negate the original slip resistance provided by the floor manufacturer because grease or grime can become ingrained within the surface of the floor
4.2 - Are any precautions taken when cleaning is being carried out?
- With some floors it is not unreasonable merely to provide warning signs that the floor is or may be wet. Typically, floors with a minimum slip resistance of 0.28 to 0.3 can be acceptable provided people are warned and only a limited number of people are likely to walk over them. If the slip resistance is less than 0.28 it is essential that the area is cordoned off and people are prevented from walking over it because even using a defensive gait they could still slip over
4.3 - Is there an emergency clean-up procedure?
- A number of the large supermarket chains have specific staff procedures for dealing with spillages and contamination. Staff are trained to look out for such problems and are instructed that first they direct customers away from the specific area and then ask a colleague to go and summon a cleaner while they remain directing customers. If no cleaner is immediately available or the spill is minor the colleague gets some absorbent paper to mop up the spill leaving the floor dry and safe. The mop and bucket approach should not be used in that situation unless the area of floor is properly cordoned off as described in 4.2
Once you have considered all twelve questions you will have a much better appreciation as to how your floor is being expected to perform. By testing the floor with the most appropriate slider materials and contaminants you can determine how the floor actually performs. From these two vital pieces of information you can then decide whether you have a problem or not. If you do have a problem, the fact that you have considered the twelve questions will almost certainly give you the basis from which you can start to solve the problem. Occasionally one is led to the conclusion that the floor is fundamentally inappropriate for that situation and the only solution is replacement or some form of remedial work to the floor to enhance its slip resistance characteristics. That, however, is the exception rather than the rule. Normally there are much more acceptable measures open - in particular those which prevent or limit the contamination reaching the floor in the first place.
So, how safe is your floor? Remember that the first and most important step in preventing slipping accidents is to know how slip resistant the floor is and to monitor it regularly to ensure it stays that way. To do that it is vital to use an instrument which gives the correct answer - be warned, most do not! For further guidance the UK Slip Resistance Group Guidelines is to be recommended, see www.ukslipresistance.org.uk. Details of SlipAlert can be found at www.slipalert.com. ?
Author Details:
Dr Bailey is Chairman of TC 339, the European Standards Committee for harmonisation of slip testing, and is the Chairman of the BSI Committee which was responsible for BS 7976, the Pendulum Standard. During the past 25 years he has tested many hundreds of floor surfaces and investigated many hundreds of slipping accidents. He is also Secretary of the UK Slip Resistance Group
Tel: +44 (0)1923 858323
Links to Products and Services
You may be interested in the following products and services on our sister site - OSE Directory.
Software, Arm and Hand Protection, Head Protection, Height Safety Equipment, Protective Clothing / Workwear, Protective Footwear, Breathing Apparatus, Respiratory Protection, Safety Knives, Showers and Eyebaths, Signs, Slip / Friction Measurement, Welding, Fabrics and Fibres, Emergency Shelters, Ear Protection, Eye Protection
