Duane Steffey

Wear and corrosion in retrieved thoracolumbar posterior internal fixation.

Study Design. Posterior thoracolumbar spine implants retrieved as part of routine clinical practice over a 2-year period were analyzed to identify wear and corrosion. Objective. Engineering analyses of retrieved posterior instrumentation for indications of performance and failure and correlation of this information with clinical factors. Summary of Background Data. Recent studies have reported spinal instrumentation particulate wear debris and have noted the importance of design considerations at implant connector interfaces. Methods. A total of 57 implants were analyzed from patients (39 female, 18 male) whose average age at implantation was 43.9 years (range, 13.7–77.4 years). Time of implantation ranged from 2 months to 13.5 years. The top 3 implantation diagnoses were radiculopathy (33%), scoliosis (30%), and back pain (25%). Metallurgical analyses were performed to characterize the wear and/or corrosion, and fractures of the implants. Results. Wear was present in 75%, corrosion in 39%, and fractures in 7% of the retrieved implants. Wear and/or corrosion was more prevalent, with respect to the total number of implants retrieved, in implants that had been in service at least 1 year. There was no evidence of corrosion in any of the Ti implants, whereas corrosion was present (with wear) in 58% of the stainless steel (SS) implants. Wear and corrosion were more frequently observed in long rods than in short rods. Implantation times were longer for SS implants than for Ti implants. Conclusions. Retrieved rods exhibited corrosion, wear, and fracture, with wear and corrosion mainly located at the interfaces with hooks, screws, or cross-connectors. The mechanisms causing this material loss in situ, as well as what local or systemic responses it may stimulate are of clinical significance and should be studied further.

Analysis of field performance using interval-censored incident data

Statistical time-to-failure analysis is a very powerful and versatile tool available to reliability engineers and statisticians for understanding and characterizing the failure risk and reliability of a component, device or system. Commonly applied methods of modeling time-to-failure involve fitting a parametric distribution, such as the Weibull probability function, using serial data on production or sales and incident data on units experiencing field failure since the launch of a product. When both the date of manufacture or sale and the date of incident are available from existing records, or can be easily ascertained by examination, the age of a failed unit can be determined exactly for purposes of analysis. Age censoring occurs, however, when one or both dates are missing—e.g., due to incomplete incident reporting or failure-induced physical damage to the unit. Excluding cases with incomplete date records involves a potentially significant loss of information in the time-to-failure analysis. We present several case studies to demonstrate how, in practice, such incidents can be treated as interval-censored observations in the time-to-failure analysis. Further, we evaluate the sensitivity of inferences to the inclusion of partially documented incidents to assess the value of this approach in practical applications.

Incorporating product retirement in field performance reliability analysis

Statistical time-to-failure analysis is a very powerful and versatile analytical tool available to reliability engineers and statisticians for understanding and communicating the failure risk and reliability of a component, device, or system. The typical approach to characterizing time to failure involves fitting a parametric distribution, such as a Weibull probability function, using time series data on sales and records of failure incidents since the launch of a product. Surviving units are treated as having right-censored failure times. If all sold products never retire from service or the retirement rate is so low that it can be ignored, the quantity of surviving units at each age point can be simply derived from the corresponding monthly sales quantity by subtracting the number of failed units sold in that month. However, modern consumer electronic products retire at much faster rates than traditional products such as coffee grinders or audio receivers. For example, most cell phone and consumer printer users upgrade to newer generations of devices even though their older devices work just fine. Properly accounting for the retirement rate in estimating the size of the at-risk population has become one of the most important steps in the field performance reliability analysis of consumer products. Neglecting to account for the shortened age of early retired units can lead to inaccurate characterization of the time-to-failure distribution and an inaccurate estimate of future failures. Retirement rates can be estimated from available information, or analyses can be performed using different assumed rates to examine the sensitivity of inferences. We present several case studies to demonstrate the practical value of accounting for the product retirement of surviving units. Results vary significantly from those obtained without addressing product retirement, thus underscoring the importance of this consideration in practical applications.

Letter to the Editor regarding Bajaj D, et al., The resistance of cortical bone tissue to failure under cyclic loading is reduced with alendronate, Bone 2014;64:57–64

• Experimental data were excluded from analyses of effects of alendronate treatment on fatigue life of bone specimens from canine ribs.

Practical applications of mixture models to complex time-to-failure data

Statistical time-to-failure analysis is a very powerful and versatile analytical tool available to reliability engineers and statisticians for understanding and communicating the failure risk and reliability of a component, device, or system. The typical approach to characterizing time to failure involves fitting a parametric distribution, such as a Weibull probability function, using historical data on sales and records of failure incidents since the launch of a product. However, such modeling assumes that each deployed unit has an equal chance of failing by any specified age. Such assumptions are often violated when two or more subpopulations exist but cannot be identified and analyzed separately. For example, production process changes, defects generated during component manufacturing, errors in the assembly process, variation of consumer behavior, and variation of operating environmental conditions can all result in significant heterogeneity in performance best described by multiple time-to-failure distributions. Available information does not always exist to separate such subpopulations. Neglecting to account for differences in time-to-failure distributions can lead to erroneous interpretations and predictions. Weibull mixture models can characterize such complex reliability data in situations when segregating subpopulations is impractical. This paper presents three case studies that successfully applied mixture modeling to field reliability data that could not be adequately modeled by standard time-to-failure distributions for homogeneous product populations.

Using ASTM E1155 to determine finished floor quality: minimum sampling requirements used to establish compliant floor flatness and levelness

A commonly accepted standard to quantify finished quality of elevated and slab-on-ground slabs is the ASTM Standard, ASTM E1155. The procedure outlines the method of profiling finished concrete to determine the flatness (FF) and levelness (FL) of the concrete finish; this method is thought of as an improvement to the “straight-edge” method and has been adopted by the American Concrete Institute (ACI, 2004) as the default measurement system for concrete finish quality. Data gathered from this standard is often used in the forensics arena to determine the root cause of finish quality issues, such as using the profiles to determine the lack of initial camber of steel beams. Although the standard is very thorough and has been used successfully for many years by owners, general contractors, and concrete subcontractors, problems arise when the standard is interpreted incorrectly and the owner is left with no guidance on the quality of the sampled data or the quality of the finished surface. This paper will explore one of the major issues in the current standard: What to do when the number of samples does not meet the required minimum number of samples. The intent and meaning of the number of sample points is not clearly defined in the standard and has been commonly misinterpreted for years, without any changes to the standard. This leaves the owner questioning the validity of the data. In a statistical conceptualization, flatness can be regarded as a characteristic (parameter) of the constructed floor, defined as the FF value one would obtain with complete measurement of the floor (infinite sampling). For any calculated FF value, regardless of sample size, hypothesis tests—or, equivalently, confidence intervals— can be constructed to determine whether the constructed floor possesses the specified flatness with a high degree of confidence. Results from simulated floor profiles adapted from a case studies show that sampling at levels below the minimum requirements established in the ASTM standard may nonetheless provide sufficient information to reliably judge the flatness of constructed floors. BACKGROUND

Risk Management of Construction Defects

Risk management of construction defects involves the collection and analysis of data to ascertain the empirical basis for alleged defects and to assess the producers or consumers risk regarding possible actions. In the context of a dispute between two or more parties, risk managers also make decisions concerning alternative options for resolution. Defects are usually identified through quality inspections and tests during construction or sometimes through the manifestation of problems after the completion of construction. Statistical sampling and estimation are used to characterize the prevalence of defects. Important considerations include the selection of a suitable sampling design, determination of an adequate sample size, and control of potential sources of bias. Defects attributable to design, materials, and method of construction may require qualitatively different approaches compared to the sampling plans implemented for defects attributable to workmanship. Repair options often vary in scope and unit cost, with implications for the associated financial risks. Keywords: alternative dispute resolution; random sampling; stratification; sample size; nonresponse bias; substitution; visual inspection; destructive testing; preponderance of evidence