Elsevier

Applied Ergonomics

Volume 44, Issue 1, January 2013, Pages 18-26
Applied Ergonomics

Carpe diem, Carpe ampulla: A numerical model as an aid to the design of child-resistant closures

https://doi.org/10.1016/j.apergo.2012.04.006Get rights and content

Abstract

The population of most developed countries is ageing. Despite continuing medical advances, ageing brings with it a host of issues, not least a loss in strength and dexterity. One major area of concern is the ability of elderly consumers to access packaged goods such as food and medicines. In previous studies, the authors developed a numerical model of a human hand that was used to investigate the effect of physical dimensions and choice of grip on joint stresses to aid the understanding between physical effort, ability and discomfort. This previous work was supported by ethnographic studies and led to recommendations for packaging design.

In this paper, a small ethnographic study is undertaken which identifies the grip types used to access to a product that is known to cause particular difficulties for the elderly, the “squeeze and turn” child-resistant closure or CRC, commonly used on medicines and cleaning products. One of the grip types used was chosen to be modelled using the numerical model developed in previous studies by the author. Model geometry and loading were adapted to simulate the “squeeze and turn” nature of the initial opening for closures of this type. A series of studies were then undertaken using different hand geometries; an average male hand, an average female hand and a fifth percentile female hand.

The prediction from the model here is that female users with smaller hands will experience more discomfort when accessing squeeze and turn CRC's and that the turn process whilst maintaining the squeeze is problematic.

Highlights

► Numerical modelling of a complex hand task is feasible. ► Higher stresses occur for smaller hand sizes. ► Simple geometry changes can be made to increase comfort.

Introduction

As the average population age in developed nations increases, issues relating to strength and dexterity become ever more important. One aspect of daily life where this is very apparent is in the design of consumer packaging. Studies have shown that many elderly people experience such difficulties in opening packaging that they will abandon products altogether (McConnell, 2004), leading to non-compliance in the case of medical packaging (de la Fuente and Bix, 2005), and possibly even malnutrition in the most vulnerable individuals. Of particular interest in the design community has been the development of a concept called ‘Inclusive’ or ‘Universal’ design, promoted by various organisations, notably the Royal College of Art in the UK. The British Standards Institute (2005) defines inclusive design as “The design of mainstream products and/or services that are accessible to, and usable by, as many people as reasonably possible...without the need for special adaptation.”

There have been numerous previous studies examining the openability of packaging across various disciplines, from the study by Rohles et al. (1983), The Department of Trade and Industry (1999), Voorbij and Steenbekkers (2002) and more recently by Su et al. (2009) and Kuo et al. (2009). The majority of this work has been in attempting to understand the accessibility of glass jars with a vacuum lug closure (VLC). This type of packaging is commonly used for sauces, preserves and pickles and the survey outlined earlier (McConnell, 2004) ranked jars after bleach bottles as the second most difficult item to open.

Yoxall et al. (2006) measured the torque strength of over 1200 people between the ages of 20 and 90 using a test instrument resembling a vacuum lug jar, shown in Fig. 1 and in use in Fig. 2. The device consists of a rotational strain gauge embedded into a glass jar. The strain gauge is removable and can be used in jars of various diameters. Similarly attached to the top of the strain sensor is a standard lid (or closure) which is again removable allowing for lids of various diameters to be tested.

Information from the strain sensor is fed to a Pico sampler and is converted to torque information through bespoke software. Instructions for the device use are provided to users via a Graphical User Interface (GUI).

The results have been presented in various studies by the authors (Yoxall et al., 2006, 2010a), and showed that for males and females there was a significant drop in strength with age, and that by 70 years of age, statistically 50% of women would not be able to access 50% of the jars they bought. Similarly the results showed that 15% of women of any age would struggle with 50% of the jars they bought; indicating that openability of jars of this type is a significant problem.

The authors have also attempted to study ease of access using analytical techniques (Yoxall and Janson, 2008). Simple mathematics showed that the limiting torque that a consumer can apply to open a container is governed by Eq. (1).Tm=μhcNArewhere, μhc is the coefficient of friction between the hand and the container cap surface, NA is the minimum human grip force required for opening and re is the external radius of the container lid.

Hence the torque a person can generate is effectively governed by six things (Rowson and Yoxall, 2011), namely:

  • 1.

    their age

  • 2.

    their gender

  • 3.

    their grip strength

  • 4.

    the coefficient of friction between the hand and the jar

  • 5.

    their wrist strength

  • 6.

    the diameter of the container.

However, the maximum torque that a person can generate may be limited by discomfort particularly for older people and those with arthritis. The authors have attempted to understand hand discomfort by using thin-film force sensors, surveys and numerical modelling of the human hand.

Yoxall et al. (2010b) looked at the forces in the hand when squeezing bottles using thin-film force sensors. These force sensors are able to produce force–time results or pressure fields. Hence these pressure maps were drawn onto schematic diagrams of a human hand. In the experiments, users were questioned about their comfort levels using the McGill's pain rating index (Melzack and Torgerson, 1971). The results showed a link between the forces in the hand and comfort when undertaking the task. Examples of the force sensor, the grips being used and schematic results of pressure on the hand are shown in Fig. 3, Fig. 4, Fig. 5.

Another approach used by the authors has been the development of numerical models to understand the forces in the hand whilst undertaking these tasks. The aim of this work has been to develop a method of understanding the relationship between discomfort and grip without the need for extensive user trials or the obvious issue of undertaking studies that lead to user discomfort (however mild).

While there are no known commercial computational tools for ergonomic hand assessment, there are several relevant studies reported in the academic literature on the subject. The majority of these take a kinematic-based approach.

Endo et al. (2006) developed a model based on kinematic data intended to assess the design of hand-held products such as mobile telephones based on how easy it is for the user to perform the movements necessary to operate the device. The model included 82 variable parameters so that it could be adapted to represent a full range of dimensional variations. The model also included algorithms to predict skin deformation and contact forces in order to assess whether or not a grasp would be stable.

Savescu et al. (2004) used a similar approach but demonstrated that the inclusion of the palmar arch is important in order to be able to predict grasping posture accurately. Yang et al. (2006) extended this work to include algorithms that predicted the reach envelope of the user and the joint torques generated in the fingers for a particular task.

Lee and Zhang (2005) used an optimisation approach to predict prehensile configurations for power-grip tasks, based on a procedure that minimised the distances from the finger joints to the surface of the object being grasped. The developed model could be used to predict grip forms for different shaped objects.

Sancho-Bru et al. (2001) also developed a dynamic model, in order to predict the free movement of the finger. It was demonstrated that inertial effects need to be included as they become significant when rapid movements are being assessed.

A somewhat different approach was taken by Rowe et al. (2005) who developed a model that evaluated the “functional demand” of a physical activity on a person. Functional demand was defined as the result of the moments created at a joint due to the external influence of performing the task and the internal moments resulting from the muscles crossing the joint and acting to counter the external moments. Quantitatively, functional demand was expressed as the ratio of the external moment to the maximum muscle moment available. Maximum muscle moments at the knee, hip, ankle and upper limbs were included in the model and a motion capture device was used to obtain joint angles and external moments occurring during tasks. This data was then imported into a bespoke software package where limb positions could be altered. The software could also accept CAD models of products for assessment. To give a visual measure of functional demand that could be easily understood by product designers, the authors used a “traffic light” system of coloured markers at each joint i.e. as the functional demand at the joint increased, the colour changed from green to amber to red.

The functional demand model of Rowe et al. (2005) can provide a similar level of information about a task. However, a hand model is not currently available and further; only the functional demand is calculated and no joint force and stress data is available.

Since no commercial models of the human hand exist, and there is a need to understand discomfort whilst undertaking hand-intensive tasks such as packaging access, the authors (Yoxall et al., 2007, 2008) developed a Finite Element model (using LS-Dyna) of a human hand. The types of grips used to open packaging such as jars and child-resistant closures, were photographed, categorised and drawn schematically (as shown in Fig. 6). Finite element models of all these grip types were then analysed (see Fig. 7a–f) with outputs for compressive and shear stresses on the joints produced for all grip types and at different grip spacings. The spherical grip model and typical output are shown in Fig. 8a and b. This was used to assess the effects of choice of grip type and hand dimensions on joint stresses during gripping tasks. Whilst understanding pain and comfort is a difficult task, assessment of stress in joints should allow for some comparative understanding.

Briefly, it was found that certain grip types and larger hand sizes result in lower joint stresses (see Fig. 9). This was supported by parallel ethnography studies which demonstrated that consumers will tend to use these grip types if the size of the packaging allows (Rowson and Yoxall, 2011).

In these previous studies, the model was used to purely assess one action; that of gripping, whether a lateral, box or spherical grip. The results showed that hand size affected the stress levels in the hand and that differing grip types produced lower stresses in the hand and therefore were in effect more comfortable than others.

From the McConell survey (2004) whilst jars ranked highly in terms of difficulty to access the highest ranked packaging product was bleach bottles with a child-resistant closure (CRC). A typical bottle and closure that uses a squeeze and turn combination are shown in Fig. 10.

Little or no previous work has been done on the ergonomic assessment of accessibility of squeeze and turn bottles (work has been undertaken however by de La Fuente and Bix (2010) on medical packaging, predominantly push and twist). Hence it was decided to use the techniques described earlier to undertake an assessment of this pack type. Therefore a series of studies are either undertaken or being developed to understand the issue in more detail, namely:

  • a device to measure with grip force and twist concurrently is currently under development

  • ethnographic studies of users accessing packaging of this type

  • analytical modelling of closure performance

  • numerical modelling of the squeeze and turn action.

This paper outlines the work undertaken in a small ethnographic study looking at users accessing this sort of packaging and the subsequent development of the numerical model for the squeeze and turn action.

Section snippets

User studies

Initially, a small ethnographic study was conducted using two different examples of child-resistant packaging. These were both bottles that featured child-resistant closures (CRCs); a “squeeze and turn” plastic bleach bottle and a “push and turn” plastic bleach bottle (see Fig. 10). The former of these uses stops on the inside of the cap and the bottle collar that prevents the cap from being twisted until it has been deformed (squeezed) far enough for the stops to clear one another. The latter

Finite element model

The numerical human hand model is described in detail in previous work by the authors (Yoxall et al., 2007, 2008). A further brief outline is given here, together with adaptations made for the specific task of modelling a child-resistant closure (CRC).

Bone geometry was created by 3D laser-scanning (ModelMaker W70, 3D Scanners, London) individual bones from a 1:1 scale skeletal right hand model (3B Scientific, Hamburg, Germany). The bones were meshed using temperature independent, constant

Results

After solving the model, plots of minimum principal stress (i.e. maximum compressive stress) and maximum shear stress were created and compared, for the timesteps at the ends of the “squeeze” and “turn” parts of the motion. Example plots of stress contours are shown in Fig. 13. In this figure the hand view is shown from underneath the closure.

The maximum compressive stress and maximum shear stress at each joint interface are shown in Table 4, normalised against the minimum value of joint

Discussion

It can be seen from Table 3 that there is a significant increase in average maximum joint interface stress when the “turn” force is applied to the closure. The joints in the index finger experience the largest increases in interface stresses. The force input does of course increase between the two timesteps of interest; however the total applied force magnitude on each finger increases by around 30% only, whereas the average maximum joint stress magnitudes increase much more significantly than

Conclusions and further work

Operations that predict lower stress will lead to lower discomfort for users and models such as the one described in the paper can be used to compare tasks. The prediction from the model here is that female users with smaller hands will experience more discomfort when accessing squeeze and turn CRC's and that the turn process whilst maintaining the squeeze is problematic. Fig. 14 showed that as hand size decreased stresses increased; therefore smaller closures would also be more comfortable for

Acknowledgements

The authors would like to thank the following individuals and organisations for their kind assistance in the preparation of this paper: Brian Walker and Miles Thornton of Arup for provision of software and modelling assistance; Tim Williams, Dirk Landheer, David Rogers and Eduardo Arvelo of Simpact Engineering Ltd. for provision of 3D laser-scan facilities and modelling assistance; The Royal Academy of Engineering and the Institution of Mechanical Engineers for travel grant awards.

References (29)

  • J. de La Fuente et al.

    User-pack interaction: insights for designing inclusive child-resistant packaging

  • L.C. Kuo et al.

    Jar-opening challenges. Part 2: estimating the force-generating capacity of thumb muscles in healthy young adults during jar-opening tasks

    Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Mechanical Engineering in Medicine

    (2009)
  • R. Melzack et al.

    On the language of pain

    Anesthesiology

    (1971)
  • Cited by (0)

    View full text