<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:wfw="http://wellformedweb.org/CommentAPI/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:sy="http://purl.org/rss/1.0/modules/syndication/" xmlns:slash="http://purl.org/rss/1.0/modules/slash/" > <channel> <title>primers | The OPEN Design Lab</title> <atom:link href="https://www.openlab.psu.edu/project_category/primers/feed/" rel="self" type="application/rss+xml" /> <link>https://www.openlab.psu.edu</link> <description>design research @ Penn State University</description> <lastBuildDate>Fri, 29 Sep 2017 00:54:53 +0000</lastBuildDate> <language>en-US</language> <sy:updatePeriod> hourly </sy:updatePeriod> <sy:updateFrequency> 1 </sy:updateFrequency> <generator>http://sites.psu.edu/?v=6.5.5</generator> <item> <title>Hybrid Models</title> <link>https://www.openlab.psu.edu/project/hybrid-models/</link> <dc:creator><![CDATA[mbp11]]></dc:creator> <pubDate>Fri, 29 Sep 2017 00:35:21 +0000</pubDate> <guid isPermaLink="false">http://sites.psu.edu/openlab/?post_type=project&p=1488</guid> <description><![CDATA[Watch “Author’s Insights: Preference” for a discussion on including anthropometry-independent preference in hybrid models. Models that are a hybrid of the boundary manikin and population model techniques are robust tools to determine appropriate adjustability of a device for a desired level of accommodation. Hybrid models often yield better results than either manikin or population techniques […]]]></description> <content:encoded><![CDATA[<div class="toolTip"><a href="http://www.youtube.com/watch?v=HqxVJQnwd34&list=UUHHXhvwvDh5wI66HvdeSUeg" target="_blank" rel="noopener">Watch “Author’s Insights: Preference”</a> for a discussion on including anthropometry-independent preference in hybrid models.</div> <p>Models that are a hybrid of the boundary manikin and population model techniques are robust tools to determine appropriate adjustability of a device for a desired level of accommodation. Hybrid models often yield better results than either manikin or population techniques alone. Hybrid models were introduced to the DfHV field in the early 2000’s, as the importance of preference and modeling real user behavior became recognized (see <a href="http://www.sae.org/technical/papers/2000-01-2172" target="_blank" rel="noopener">Reed, 2000</a> and <a href="http://openlab.psu.edu/?p=437">Parkinson, 2005</a>).</p> <p>Hybrid models combine the experimental nature of population models with the statistical nature of manikin-based design. A simple implementation of a hybrid model works like this:</p> <ol> <li>Create a prototype of the device being designed. This prototype should have ample adjustability in the dimension(s) under consideration.</li> <li>Gather a relatively small sample of experimental participants, and determine how the experiment will be conducted.</li> <li>Take relevant measurements of the sample population. Stature and BMI are easy to measure and are almost always collected, but other measurements specific to the problem may also need to be obtained.</li> <li>Run an experiment in which each person in the sample population uses the prototype. Record user’s selected setting(s).</li> <li>Create a regression model of the data that correlates users’ settings with one or more measures of their anthropometry.</li> <li>Enter appropriate limits of anthropometry (5th and 95th percentile stature, for example) into regression equation, to get limits of the device’s range of adjustment for desired level of accommodation. See figure below for a sample population, regression line, and adjustment limits in orange determined by 2.5th and 97.5th stature of the target population under consideration.</li> </ol> <p><a href="http://openlab.psu.edu/wp-content/uploads/2014/03/regression.png"><img loading="lazy" decoding="async" class="alignleft size-full wp-image-649" src="http://openlab.psu.edu/wp-content/uploads/2014/03/regression.png" alt="regression" width="300" height="192" /></a></p> <p>More advanced implementations of hybrid models use not only the linear component of the regression, but also retain a measure of scatter (e.g., root mean squared error, RMSE). This measure of scatter is used to add a stochastic component to the simulations that represents user preference. An implementation of a hybrid model works like this (starting with step 5 above):</p> <ol> <li>Create a virtual population of a large number of people (e.g., 1000). Randomly sampling the given number of people from an existing anthropometric database (e.g., ANSUR) is one way to accomplish this.</li> <li>Using the regression equation from step 5 above, determine the preferred device setting for each person. This equation should include not only the mean location predicted by the slope and intercept of the regression, but should also add a random component generated by sampling random values from a normal distribution with standard deviation equal to the RMSE of the regression.</li> <li>Determine limits for the design by selecting cutoffs that accommodate the desired percentage of target users. For example, to accommodate 95% of users, set the device adjustment cutoffs so that the central 950 of 1000 users are accommodated. See figure below for a sample virtual population of 1000 people created from the regression above, and adjustment limits in orange to accommodate the central 950.</li> </ol> <p><a href="http://openlab.psu.edu/wp-content/uploads/2014/03/virtualpop.png"><img loading="lazy" decoding="async" class="alignleft size-full wp-image-650" src="http://openlab.psu.edu/wp-content/uploads/2014/03/virtualpop.png" alt="virtualpop" width="300" height="196" /></a></p> <p>While hybrid models offer more accurate results than manikin or population models, they require both an experiment and statistical analysis, and therefore cost more time and money. See <a href="http://openlab.psu.edu/?p=391">Garneau, 2007</a> for a more thorough discussion of this procedure and a sample case study of the method applied to the design of an exercise seat cycle.</p> ]]></content:encoded> <post-id xmlns="com-wordpress:feed-additions:1">1488</post-id> </item> <item> <title>Population Models</title> <link>https://www.openlab.psu.edu/project/population-models/</link> <dc:creator><![CDATA[mbp11]]></dc:creator> <pubDate>Fri, 29 Sep 2017 00:33:47 +0000</pubDate> <guid isPermaLink="false">http://sites.psu.edu/openlab/?post_type=project&p=1486</guid> <description><![CDATA[Watch “Fundamentals of DfHV: Experiments” for a summary on setting up experiments for task-oriented models. A traditional spatial analysis design tool that is an alternative to the boundary manikin approach is the task-oriented percentile model. Such a model makes use of tests involving a sample population performing a task related to the dimension under consideration. […]]]></description> <content:encoded><![CDATA[<div class="toolTip"><a href="http://www.youtube.com/watch?v=blVoM8T2UsU" target="_blank" rel="noopener">Watch “Fundamentals of DfHV: Experiments”</a> for a summary on setting up experiments for task-oriented models.</div> <p>A traditional spatial analysis design tool that is an alternative to the boundary manikin approach is the task-oriented percentile model. Such a model makes use of tests involving a sample population performing a task related to the dimension under consideration. Required adjustability is then defined by the selections or capabilities of the desired proportion of users. Both a sufficiently large representative sample population and a workable prototype are required. For example, this approach forms the basis for the SAE International recommended practices, which are used for vehicle design.</p> <p>These models are an improvement on manikin-based approaches in some ways, since they specifically model the outcome measure of interest (e.g., reach, eye location, driver-selected seat position), rather than trying to predict the population distributions of those outcomes from boundary cases defined by anthropometry. However, they require extensive human-subject data from a similar task scenario and they are essentially univariate, dealing with only a single outcome measure (e.g., reach) at one time. Population models are not easily adapted to other conditions, tasks, or populations; they may quickly become outdated as these qualities change over time.</p> <p>Hybrid models expand on the population model approach by allowing the model to be extrapolated to populations different than the one from which the data were gathered. Hybrid models combine many of the virtues of the boundary manikin and population model approaches. Implementation of hybrid models is discussed on the <a href="http://openlab.psu.edu/?p=648">Hybrid Models</a> guidelines page.</p> ]]></content:encoded> <post-id xmlns="com-wordpress:feed-additions:1">1486</post-id> </item> <item> <title>Proportionality Constants</title> <link>https://www.openlab.psu.edu/project/proportionality-constants/</link> <dc:creator><![CDATA[mbp11]]></dc:creator> <pubDate>Fri, 29 Sep 2017 00:30:42 +0000</pubDate> <guid isPermaLink="false">http://sites.psu.edu/openlab/?post_type=project&p=1483</guid> <description><![CDATA[Proportionality Constants were one of the earliest methods developed to predict human anthropometry. They are typically calculated by taking a large sample of anthropometric data and determining either the mean or 50th percentile ratio of the length of each measure of interest to stature. Drillis and Contini were among the first to publish mathematical ratios […]]]></description> <content:encoded><![CDATA[<p>Proportionality Constants were one of the earliest methods developed to predict human anthropometry. They are typically calculated by taking a large sample of anthropometric data and determining either the mean or 50th percentile ratio of the length of each measure of interest to stature. Drillis and Contini were among the first to publish mathematical ratios of many body dimensions to stature (Drillis and Contini, 1966). See figure below for a simplified diagram of Drillis and Contini’s proportionality constants.</p> <p><a href="http://openlab.psu.edu/wp-content/uploads/2014/03/drillis-contini.png"><img loading="lazy" decoding="async" class="alignleft size-full wp-image-642" src="http://openlab.psu.edu/wp-content/uploads/2014/03/drillis-contini.png" alt="drillis-contini" width="240" height="303" srcset="https://www.openlab.psu.edu/files/2014/03/drillis-contini.png 240w, https://www.openlab.psu.edu/files/2014/03/drillis-contini-238x300.png 238w" sizes="(max-width: 240px) 100vw, 240px" /></a></p> <p>These values have been extensively used as a design tool because they provide a means of estimating the lengths of many body segments while knowing only the stature of an individual. An example design process using proportionality constants works like this:</p> <ol> <li>Determine which body dimension can be used to most accurately predict adjustability levels. For example, trochanteric height (leg length) may be used to predict seat height adjustability range on a stationary exercise bike.</li> <li>Determine the cutoff percentiles for the desired accommodation level for the artifact. If 95% accommodation is desired, the 2.5th and 97.5th percentiles are used.</li> <li>Find the accommodation range. To do this, multiply the proportionality constant of the body dimension determined in step 1 by the stature corresponding to the low cutoff percentile determined in step 2. This provides the smallest body dimension length that will be accommodated. Multiply the same proportionality constant by the stature corresponding to the high cutoff percentile to obtain the largest body dimension that will be accommodated. The difference between these two lengths is the accommodation range.</li> </ol> <h3>Boundary Ratios</h3> <p>While simple to use, proportionality constants have many limitations. Those from Drillis and Contini were calculated from a population that has never been validated, and they did not provide formal definitions of the body dimensions each ratio predicted. Furthermore, proportionality constants are used under the false assumption that an nth percentile person by stature is composed of nth percentile body segments. A recent study (see <a href="http://openlab.psu.edu/?p=408">Fromuth, 2008</a>) has overcome the above drawbacks by developing a set of Boundary Ratios. There are however, drawbacks to all proportionality constant models that cannot be overcome. In DfHV, posture and user preference heavily influence accommodation levels and these often do not correlate with anthropometry. Proportionality constant models assume that two people of the same size will interact with a designed artifact in the same way. Due to these limitations, it is best to gather data from a study of the target population to obtain more accurate estimates of accommodation levels using either population or hybrid models.</p> ]]></content:encoded> <post-id xmlns="com-wordpress:feed-additions:1">1483</post-id> </item> <item> <title>Formula SAE</title> <link>https://www.openlab.psu.edu/project/formula-sae/</link> <dc:creator><![CDATA[mbp11]]></dc:creator> <pubDate>Fri, 29 Sep 2017 00:29:43 +0000</pubDate> <guid isPermaLink="false">http://sites.psu.edu/openlab/?post_type=project&p=1481</guid> <description><![CDATA[Design guidelines for Formula SAE teams.]]></description> <content:encoded><![CDATA[<p>Design guidelines for Formula SAE teams.</p> ]]></content:encoded> <post-id xmlns="com-wordpress:feed-additions:1">1481</post-id> </item> <item> <title>Digital Human Models</title> <link>https://www.openlab.psu.edu/project/digital-human-models/</link> <dc:creator><![CDATA[mbp11]]></dc:creator> <pubDate>Fri, 29 Sep 2017 00:27:41 +0000</pubDate> <guid isPermaLink="false">http://sites.psu.edu/openlab/?post_type=project&p=1479</guid> <description><![CDATA[Three-dimensional manikins are also known as Digital Human Models (DHM). These are software representations of humans that enable designers to visualize the effectiveness of a design before a physical prototype is constructed. DHM programs such as Jack, Ramsis, and Safework are derived from the same types of technology as Computer-Aided Design (CAD) programs and actually […]]]></description> <content:encoded><![CDATA[<p>Three-dimensional manikins are also known as Digital Human Models (DHM). These are software representations of humans that enable designers to visualize the effectiveness of a design before a physical prototype is constructed. DHM programs such as Jack, Ramsis, and Safework are derived from the same types of technology as Computer-Aided Design (CAD) programs and actually allow users to import their 3D CAD models into a virtual environment. Then, DHMs of various sizes can be placed into this environment along with the model for design analysis.</p> <p>The DHM programs can then be used to assess many design concerns. For example, automotive companies can utilize DHM to examine if the current seat adjustability will allow a wide range of users to reach all of the needed controls. DHM not only benefits the end user, it is also incredibly useful for examining manufacturability or maintenance situations where individuals will need to be able to reach components for either assembly or repair. Furthermore, DHM can be used to optimize workplace or workstation design to reduce health or safety concerns. Being able to do all of this on a computer rather than using a physical prototype results in faster, higher quality, and more accessible designs that also lower cost.</p> <p><a href="http://openlab.psu.edu/wp-content/uploads/2014/03/DHMb.png"><img loading="lazy" decoding="async" class="alignleft size-medium wp-image-635" src="http://openlab.psu.edu/wp-content/uploads/2014/03/DHMb-300x204.png" alt="DHMb" width="300" height="204" /></a></p> <p>While DHM can be useful as a validation tool (e.g., for performing clearance checks), it is not an “end-all” solution. One of its major downfalls is that DHM does not include any of the preference considerations that are a main focus of DfHV. The impressive visuals rendered by DHM software can enhance the communication of design objectives, but they may also serve to obfuscate the limitations of DHM’s univariate assumptions (i.e., manikins are created via anthropometric scaling). Nonetheless, they are an incredibly useful tool and, if used wisely, can be beneficial to the design process.</p> ]]></content:encoded> <post-id xmlns="com-wordpress:feed-additions:1">1479</post-id> </item> <item> <title>Boundary Manikins</title> <link>https://www.openlab.psu.edu/project/boundary-manikins/</link> <dc:creator><![CDATA[mbp11]]></dc:creator> <pubDate>Fri, 29 Sep 2017 00:26:06 +0000</pubDate> <guid isPermaLink="false">http://sites.psu.edu/openlab/?post_type=project&p=1477</guid> <description><![CDATA[“Manikins” are typically two- or three-dimensional representations of the human form with external contours intended to represent human body size and shape for design. They exist as 2D templates and as 3D computerized manikins (e.g., Jack). Digital Human Modeling, or DHM, refers to the use of computerized manikins for design. A “boundary manikin” or “boundary […]]]></description> <content:encoded><![CDATA[<div id="content_opl"> <section> <article id="post-632"> <section> <div> <p>“Manikins” are typically two- or three-dimensional representations of the human form with external contours intended to represent human body size and shape for design. They exist as 2D templates and as 3D computerized manikins (e.g., Jack). <a href="http://openlab.psu.edu/?p=634">Digital Human Modeling</a>, or DHM, refers to the use of computerized manikins for design. A “boundary manikin” or “boundary case” refers to a body geometry that lies at the limit of acceptability. For a design problem in which only one body dimension is relevant and both “small” and “large” people must be considered, only two cases are typically considered to describe the upper and lower limits of acceptability. For example, to accommodate 95% of the population, one might use the 2.5th-percentile and 97.5th-percentile values of the measure of interest as boundary cases. Note that since the distribution of body sizes is continuous, the specific level of accommodation (95%) could be achieved by targeting any number of segments. Generally the range is selected to minimize the amount of adjustability or material and cost required (e.g., the 0th to 95th percentile or the 2.5th to 97.5th percentile portion of the distribution).</p> <p>Since stature and weight data are the most easily obtained and therefore prevalent in databases, distributions of those variables are often used to determine the sizes of the “small” and “large” virtual users. To position a manikin, it is helpful to have knowledge of the length of the relevant dimension(s). This dimension may not be stature or weight, but instead might be leg length or shoulder height, for example. Therefore, in the event that the true measure of interest is something besides stature, <a href="http://openlab.psu.edu/design-tools-boundary-manikins/proportionality_constants.php">proportionaility constants</a> are often used. These represent the average length of a particular body segment as a proportion of stature. This approach is imprecise, however, since there is no “standard” person with all dimensions belonging to the same percentile. For example, when a manikin representing the body as a kinematic linkage is scaled so that an overall dimension, such as stature, meets some target percentile, the body dimensions that make up the aggregate dimension do not themselves define useful design limits. That is, a person who is 5th-percentile by stature has other body dimensions that vary widely from the 5th-percentile for those measures.</p> <p>The <a href="http://openlab.psu.edu/tools/calculators/AnsurDimensionSelect.php" target="_blank" rel="noopener">ANSUR database</a> is a useful source of anthropometric data for configuring boundary manikins because it contains hundreds of body segment measurements collected from thousands of people. The ANSUR database tool on this website aids in applying ANSUR data. However, caution must be exercised in the application of this data since the ANSUR population does not itself define a representative target population (the data was collected on the military population in 1988). <a href="http://www.cdc.gov/nchs/nhanes.htm" target="_blank" rel="noopener">NHANES data</a> are another useful source of anthropometric and demographic data. NHANES data are collected every few years, so the measures are more representative of the general US population. However useful measures for spatial design are mostly limited to stature and BMI. Proportionality constants may be used to calculate the actual measure of interest. Further details regarding sources of anthropometric data may be found in the <a href="http://openlab.psu.edu/?p=621">Anthropometric Database</a> guidelines on this site.</p> <p>Boundary manikins have various limitations when used for design. Some of these limitations are addressed in the publications on this site. Examples are <a href="http://openlab.psu.edu/?p=391">Garneau and Parkinson (2009, JMD)</a> and <a href="http://openlab.psu.edu/?p=347">Garneau and Parkinson (2009, JED)</a>, which provides a more general discussion of the appropriateness of several design methods.</p> </div> </section> </article> </section> </div> ]]></content:encoded> <post-id xmlns="com-wordpress:feed-additions:1">1477</post-id> </item> <item> <title>Anthropometric Databases</title> <link>https://www.openlab.psu.edu/project/anthropometric-databases/</link> <dc:creator><![CDATA[mbp11]]></dc:creator> <pubDate>Fri, 29 Sep 2017 00:20:49 +0000</pubDate> <guid isPermaLink="false">http://sites.psu.edu/openlab/?post_type=project&p=1474</guid> <description><![CDATA[Numerous organizations around the world routinely conduct anthropometric surveys of different populations (e.g., military, civilian, etc.) and organize the information into databases. In the surveys that look to accurately represent the compositions of these populations (henceforth referred to as “reference populations”), the subjects are sampled based on demographic variables such as their age, gender, race, […]]]></description> <content:encoded><![CDATA[<p>Numerous organizations around the world routinely conduct anthropometric surveys of different populations (e.g., military, civilian, etc.) and organize the information into databases. In the surveys that look to accurately represent the compositions of these populations (henceforth referred to as “reference populations”), the subjects are sampled based on demographic variables such as their age, gender, race, ethnicity, etc. The detailed anthropometric and demographic information and the large amounts of information contained in the databases make them valuable design tools.</p> <p>Designers may employ either of the following two approaches when using an anthropometric database:</p> <p><a href="http://openlab.psu.edu/wp-content/uploads/2014/03/databases_stat-bmi-plot.png"><img loading="lazy" decoding="async" class="alignnone size-medium wp-image-626" src="http://openlab.psu.edu/wp-content/uploads/2014/03/databases_stat-bmi-plot-233x300.png" alt="databases_stat-bmi-plot" width="233" height="300" /></a></p> <ul> <li>Performing accommodation analyses directly on the reference population and extending the results to the target user population. This approach is faulty on three counts: a) it requires assuming that the reference and target populations are similarly composed, as measured by the distributions of demographic variables, b) it neglects the impact of temporal changes in the reference population anthropometry, and c) it fails to consider other reasons (e.g., high fitness levels and the absence of pregnant women in military populations) for possible differences in anthropometric distributions. The figures illustrate the differences in the compositions of three databases.</li> <li>Utilizing various techniques (e.g., principal components analysis, the regression with residual variance methodology) to extrapolate the relationships found to exist in the reference population anthropometry to the target user population. Doing so allows for the estimation of user population anthropometry; accommodation analyses may be carried out on these estimates.</li> </ul> <p><a href="http://openlab.psu.edu/wp-content/uploads/2014/03/databases_demography.png"><img loading="lazy" decoding="async" class="alignnone size-medium wp-image-628" src="http://openlab.psu.edu/wp-content/uploads/2014/03/databases_demography-300x122.png" alt="databases_demography" width="300" height="122" srcset="https://www.openlab.psu.edu/files/2014/03/databases_demography-300x122.png 300w, https://www.openlab.psu.edu/files/2014/03/databases_demography.png 537w" sizes="(max-width: 300px) 100vw, 300px" /></a></p> <p>Three anthropometry databases are frequently used in research in the OPEN Design Lab:</p> <ul> <li>NHANESThe focus of the three National Health Examination Surveys (NHES) was the civilian population of the United States from 1959 to 1970. NHES was succeeded in 1971 by the National Health and Nutrition Examination Survey (<a href="http://www.cdc.gov/nchs/nhanes.htm" target="_blank" rel="noopener">NHANES</a>) run by the Centers for Disease Control and Prevention (<a href="http://www.cdc.gov/" target="_blank" rel="noopener">CDC</a>). NHANES I, II, and III were compiled in the years 1971-1975, 1976-1980, and 1988-1994, respectively. Of late (1999-2006), NHANES databases have been compiled every two years in order to avoid temporal errors in anthropometric analyses. However, they are not comprehensive, and are characterized by a dearth of anthropometry in addition to stature and BMI.</li> <li>ANSURThe 1988 U.S. Army Anthropometry Survey (ANSUR) is widely used, thanks to: a) the large number of anthropometry contained in it and b) the use of techniques such as population oversampling and statistical matching in its creation, something that allows for future army populations to be simulated by adjusting the demographic factors as required. See <a href="http://www.dtic.mil/dticasd/docs-a/anthro_military.html" target="_blank" rel="noopener">the DTIC website for this data.</a></li> <li>CAESARThe Civilian American and European Surface Anthropometry Resource (CAESAR) is the first database to contain data from 3-D body scans in addition to conventional 1-D measures. The subjects are random North American and European volunteers, so CAESAR is not representative of any particular population. (See Blackwell, S., Brill, T., Boehmer, M., Fleming, S., Kelly, S., Hoeferlin, D., Robinette, K., 2008, “CAESAR Survey Measurement and Landmark Descriptions”, Air Force Research Laboratory, Wright-Patterson AFB, OH)</li> </ul> <p>Other surveys from around the world include Germany’s MikroZensus, Japan’s Human Engineering for Quality of Life surveys, and the Health Survey for England. The use of these databases is simplified by numerous tools that are designed to process anthropometric information and present it in more readily-usable forms. Examples of such tools are <a href="http://www.myanthro.com/" target="_blank" rel="noopener">myAnthro</a>, a source for NHANES and ANSUR data, and PeopleSize 2008 (<a href="http://www.openerg.com/psz/" target="_blank" rel="noopener">Open Ergonomics Ltd.</a>), which generates anthropometric estimates for different target user populations.</p> ]]></content:encoded> <post-id xmlns="com-wordpress:feed-additions:1">1474</post-id> </item> </channel> </rss>