Optimizing the fuzzy-nets training scheme using the Taguchi parameter design

Fuzzy nets have been proposed to combine the learning ability of neural networks and the reasoning ability of fuzzy logic to deal with complex control systems. This paper presents a systematic way of identifying the significant factors and optimising the performance of a fuzzy-nets application. To present the methodology, a model of a truck backing up has been evaluated. Four factors were considered:

1. The number of training sets.
2. The number of fuzzy regions.
3. The membership functions.
4. The fuzzy reasoning methods which would affect the performance of the fuzzy-nets training scheme in nonlinear applications.

The Taguchi parameter design was implemented with anL ₉ (3⁴) orthogonal array to identify the optimal combination for training consideration. Both raw and signal-to-noise (S/N) ratios were evaluated to identify the optimal combination for the performance of fuzzy-nets training with very limited variation. The performance of the proposed fuzzy-nets scheme for the model of the truck backing up was represented by the average errors between the truck and loading dock: 0.178 units and 0.204 degrees. The results demonstrate that the Taguchi parameter design is a robust approach for optimising the performance of the fuzzy-nets training scheme.     

Optimizing surface finish in a turning operation using the Taguchi parameter design method

This paper presents an application of the Taguchi parameter design method to optimizing the surface finish in a turning operation. The Taguchi parameter design method is an efficient experimental method in which a response variable can be optimized, given various control and noise factors, and using fewer experimental runs than a factorial design. The control parameters for this operation included: spindle speed, feed rate, depth of cut, and tool nose radius. Noise factors included varying room temperature, as well as the use of more than one insert of the same specification, which introduced tool dimension variability. A total of 36 experimental runs were conducted using an orthogonal array, and the ideal combination of control factor levels was determined for the optimal surface roughness and signal-to-noise ratio. A confirmation run was used to verify the results, which indicated that this method was both efficient and effective in determining the best turning parameters for the optimal surface roughness.     

The effect of the Rim Cutter on cement pressurization and penetration on cemented acetabular fixation in total hip arthroplasty: an in vitro study

The Rim Cutter (Stryker Orthopedics, Mahwah, New Jersey) is a tool designed to cut a ledge inside the rim of the acetabulum, onto which a precisely trimmed, cemented, flanged cup can be fitted. The aim was to investigate the effect of the Rim Cutter on the intra-acetabular cement mantle pressure and the depth of cement penetration during cup insertion. The study had two parts. In the first part, hemi-pelvis models were fitted with pressure sensors. Pressure in the acetabulum was measured on insertion of a conventional cemented flanged cup with and without the use of a Rim Cutter to prepare the rim of the acetabulum. The second part assessed cement penetration when the same cups were inserted into a foam shell model. The shell was mounted in a jig and had holes drilled in it; the distance that cement penetrated into the holes was measured. A significant increase in cement pressure at the apex (p = 0.04) and the rim (p = 0.004) is seen when the Rim Cutter is used. Cement penetration in the Rim Cutter group was significantly increased at the rim of the acetabulum (p = 0.003). Insertion of a flanged cup after the acetabulum is prepared with the Rim Cutter leads to a significant increase in cement pressure and penetration during cup insertion in vitro when compared with conventional flanged cups.     

A hip simulator study of metal-on-metal hip joint device using acetabular cups with different fixation surface conditions

In vitro wear data for hip joint devices reported in the literature vary in a wide range from one simulator study to another sometimes for the same type of device tested under identical physiological testing conditions. We hypothesized that non-bearing surface condition of the testing components could be an important factor affecting the simulator wear results. To confirm this hypothesis, fifteen 50 mm metal-on-metal hip resurfacing devices with identical bearing specifications were tested in a ProSim hip wear simulator for 5 million cycles. The heads were standard Birmingham Hip Resurfacing (BHR) heads; whilst the pairing acetabular cups were identical to the standard BHR cup except their different back surface conditions, including: (a) off-the-shelf products after removing the hydroxyapatite (HA) coating; (b) semi-finished products without HA coating; and (c) purposely-made cups without cast-in beads and HA coating. Results showed that the different back surfaces of the cups used indeed caused significantly large variations in the gravimetrically measured wear loss. We postulated that materials loss from the non-bearing surface of the testing components could contribute to the gravimetrically measured wear loss during a wear simulator test both directly and indirectly. The results presented in this paper pertain to In vitro wear simulator study and have little clinical relevance to the performance of any implant in vivo.     

The contribution of the micropores in bone cement surface to generation of femoral stem wear in total hip replacement

Although cemented total hip replacement has long been recognized as a situation that can lead to wear, the wear generated on the femoral stem has not been well documented, especially with regard to how this wear is initiated and propagated. This present work aimed to further investigate this issue based on a comprehensive study on surface morphology of the femoral stem and the bone cement, which were collected from seven in vitro wear simulations. It was shown that the wear locations on the stem surface compared well with the results of retrieval studies, and the boundaries of the worn areas matched well the edges of the micropores present in the bone cement surface. This indicated that the micropores could potentially contribute to the generation of femoral stem wear. In addition, metallic debris was detected around the micropores from the simulation with increased loading cycles. However, no evidence of macro-cracks was observed across the cement mantle in spite of the presence of micro-cracks initiated at the edge of the micropores. This study demonstrated a possible cause for progression of femoral stem wear and it may have an important bearing on the long term durability of cemented hip prosthesis.     

Optimal parameter design for chip-on-film technology using the Taguchi method

Chip on film (COF) is a new technology aggressively developed by liquid crystal display (LCD) module manufacturers. COF is a potential replacement method for tape-automated bonding (TAB) and chip-on-glass (COG) technology. COF technology has the following advantages: low power consumption, low cost, small structure, light weight and high resolution. It has become the major LCD module application. However, COF technology has several nonconforming items that should be reduced. These items include bonding misalignment, insufficient bonding strength and surface mount technology (SMT), solder joint defects, etc. The purpose of this study is to determine the optimum process parameters for COF technology improvement. An L₁₈ orthogonal array is employed to conduct the design of experiment for finding the optimum process parameters. The results show that bonding misalignment is reduced and bonding strength increased .     

Axisymmetric mechanical analysis of ceramic heads for total hip replacement

An axisymmetric, mechanical analysis of conical press-fit ceramic heads is performed. The head strength and its fracture modes are assessed experimentally. The stress field is examined by finite element, strain gauge and photoelastic methods. An alternative head design, characterized by a cylindrical engagement with the stem, is analysed with the same techniques and its merits are explored.     

Laboratory wear tests and clinical observations of the penetration of femoral heads into acetabular cups in total replacement hip joints: I: Charnley prostheses with polytetrafluoroethylene acetabular cups

A quarter of a century has now elapsed since the late Professor Sir John Charnley introduced his low friction arthroplasty of the hip based on a stainless steel femoral component and an unfilled polytetrafluoroethylene (PTFE) acetabular cup. The advantages of the very low friction resulting from the use of PTFE and a femoral head of diameter 22 mm were unfortunately outweighed by a very rapid penetration of the femoral head into the acetabular cup. Charnley abandoned the use of PTFE after some 3 years and 300 operations and subsequent measurements revealed a penetration rate of 2.26 mm year^?1. In 1962 he adopted ultrahigh molecular weight polyethylene as the socket material and this material is now used almost exclusively in total replacement joints.In this paper we report further laboratory wear studies of PTFE on stainless steel in disc machines and compare the laboratory wear factors with those deduced from clinical observations. Charnley's clinical penetration rates have been analysed in terms of Paul's loading cycles for steady walking and measurements of the walking activity of elderly patients by Wallbridge and Dowson.It is shown that the wear factor for PTFE and stainless steel is about 3 × 10^?5mm³N^?1m^?1 and that there is excellent agreement between the mean value obtained from a variety of laboratory experiments involving environments of atmospheric air, distilled water, bovine synovial fluid and Ringers' solution with various counterface roughnesses and the wear factors derived from clinical observations.     

New joints for the Millennium: wear control in total replacement hip joints

Hip joint replacement is described as the greatest achievement in orthopaedic surgery in the twentieth century. The field has been dominated for some forty years by implants based upon metallic femoral heads and stems and polymeric acetabular cups. At the dawn of the new Millennium, many alternative materials and designs are now being proposed or evaluated. The reasons for these developments and the current contributions of engineering science and tribology to advances in hip replacement are discussed. Illustrations are presented of the significant changes being proposed or introduced. While the new designs of total hip replacements offer exciting engineering contributions to the future of joint replacement, the long-term benefits to patients will depend upon the biological response to the new devices.     

Long-term implant-bone fixation of the femoral component in total knee replacement

Success of total knee replacement (TKR) depends on the prosthetic design. Aseptic loosening of the femoral component is a significant failure mode that has received little attention. Despite the clinical relevance of failures, no protocol is available to test long-term implant-bone fixation of TKR in vitro. The scope of this work was to develop and validate a protocol to assess pre-clinically the fixation of TKR femoral components. An in vitro protocol was designed to apply a simplified but relevant loading profile using a 6-degrees-of-freedom knee simulator for 1,000000 cycles. Implant-bone inducible micromotions and permanent migrations were measured at three locations throughout the test. After test completion, fatigue damage in the cement was quantified. The developed protocol was successfully applied to a commercial TKR. Additional tests were performed to exclude artefacts due to swelling or creep of the composite femur models. The components migrated distally; they tilted towards valgus in the frontal plane and in extension in the sagittal plane. The migration patterns were consistent with clinical roentgen-stereophotogrammetric recordings with TKR. Additional indicators were proposed that could quantify the tendency to loosen/stabilize. The type and amount of damage found in the cement, as well as the migration patterns, were consistent with clinical experience with the specific TKR investigated. The proposed pre-clinical test yielded repeatable results, which were consistent with the clinical literature. Therefore, its relevance and reliability was proved.     

Morphological optimization of electrospun polyacrylamide/MWCNTs nanocomposite nanofibers using Taguchi’s experimental design

The morphological characteristic of electrospun polyacrylamide/multi-walled carbon nanotube (PAAm/MWCNTs) nanocomposite nanofibers is optimized in this work using Taguchi’s experimental design. The optimization is performed considering the effect of PAAm concentration, MWCNTs content, flow rate, and applied voltage on average nanofibers diameter. The reasonable dispersion of MWCNTs in PAAm solution is first ascertained via optical microscopy method. The experimental data required for the optimization process are then provided by statistical calculations on field-emission scanning electron microscopy images of the samples formulated based on a designed L ₉ orthogonal array. PAAm concentration is found to have the most contribution on final fibers morphology according to the results obtained from simultaneous implementation of the analysis of variance and mean effect assessment. Therefore, PAAm concentration, which is in consistence with solution viscosity and surface tension parameter, is found to have the most contribution to forming nanofibers including the finest fiber diameter. On the contrary, the flow rate of solution among the selected parameters shows the least effect on average nanofiber diameter.     

Investigation into the tribological condition of acetabular tissue after bipolar joint replacement hip surgery

The aim of this research was to investigate the tribological condition of acetabular tissue before and after bipolar hip surgery. Articular cartilage was taken from the femoral head of patients undergoing primary joint replacement as a control. Tissue was also taken from the acetabular cups of patients undergoing revision hip surgery after primary bipolar surgery and compared with the control cartilage. The biomechanical characteristics of the two tissue types were tested using friction and compression tests. The friction tests were carried out on a sliding friction rig under nominal contact stresses of 0.5 and 4 MPa. The compression tests were carried out under a 0.8 MPa contact stress. The majority of the bipolar patients produced friction coefficients that were significantly higher than those produced by the control group, and the compression tests highlighted that the tissue from the bipolar patients produced a much greater rate of increase in displacement compared with the control cartilage. Histology showed major differences between the control cartilage and the bipolar tissue. The control cartilage showed a healthy collagen structure with a good distribution of proteoglycan whereas the majority of the bipolar tissue had lost tissue architecture and had a sparse fibrous structure. The high friction coefficients with the bipolar tissue imply that the frictional torque at the outer head of the bipolar prosthesis would be large compared with the inner bearing frictional torque. It was therefore predicted that the motion of the bipolar prosthesis should occur at the inner bearing.     

Radiographic assessment of the cement mantle thickness of the femoral stem in total hip replacement: a case study of 112 consecutive implants

The results are reported of a radiographic study of cement mantle thickness in 112 consecutive primary hip replacements. Measurements were made by three observers of the apparent cement thickness medially and laterally using standard anterior-posterior radiographs. The average cement thickness was 3.2 mm, which is 1.2 mm greater than the size difference between the broach and the prosthesis, and was in the range 2-5 mm in 67 per cent of all measurement points. This has significance for the design of instrumentation to prepare the femoral cavity to give a defined cement mantle thickness. There was a greater cement mantle thickness proximally than distally. In 95 cases it was possible to determine the orientation of the stem within the cement mantle, which showed an even distribution between varus and valgus orientation; 49 per cent were within 1 degree of neutral and only one case was more than 5 degrees from neutral.     

Reproduction of fretting wear at the stem-cement interface in total hip replacement

The stem-cement interface experiences fretting wear in vivo due to low-amplitude oscillatory micromotion under physiological loading, as a consequence it is considered to play an important part in the overall wear of cemented total hip replacement. Despite its potential significance, in-vitro simulation to reproduce fretting wear has seldom been attempted and even then with only limited success. In the present study, fretting wear was successfully reproduced at the stem-cement interface through an in-vitro wear simulation, which was performed in part with reference to ISO 7206-4: 2002. The wear locations compared well with the results of retrieval studies. There was no evidence of bone cement transfer films on the stem surface and no fatigue cracks in the cement mantle. The cement surface was severely damaged in those areas in contact with the fretting zones on the stem surface, with retention of cement debris in the micropores. Furthermore, it was suggested that these micropores contributed to initiation and propagation of fretting wear. This study gave scope for further comparative study of the influence of stem geometry, stem surface finish, and bone cement brand on generation of fretting wear.     

A simulator study of friction in total replacement hip joints.

Frictional behaviour of 22 different femoral head-acetabular cup combinations was studied on a new servo-hydraulic microcomputer-controlled hip joint simulator using various flexion-extension angle and superior-inferior load set value waveforms and using distilled water at 37 +/- 1 degrees C as lubricant. Six different head materials were included in the study, whereas all cups were ultra-high molecular weight polyethylene (UHMWPE). Most head-cup combinations studied are commercially available. No distinctly superior joint design can be pointed out, but the frictional behaviour of alumina ceramic against UHMWPE proved overall most favourable (mu min was 0.02), whereas that of non-ion-implanted titanium alloy Ti-6Al-4V against UHMWPE proved strikingly poor (mu max was 0.15). The lowest frictional torque was in 22 mm joints, but frictional torque did not always increase straightforwardly with increasing diameter of the femoral head. The measurements form an extensive comparison between a wide variety of head-cup combinations. The simulator is apparently a useful instrument in the study of frictional behaviour of new designs, materials, surface treatments and coatings that are frequently introduced.     

Predictive role of the lambda ratio in the evaluation of metal-on-metal total hip replacement

The wear of metal-on-metal bearings is affected by various design parameters, such as the clearance or surface roughness. It would be very useful to have a significant indicator of wear according to these design parameters, such as the lambda ratio. Three different batches of cast high- and low-carbon cobalt-chromium hip implants (28 mm, 32 mm, and 36 mm diameters) were tested in a hip joint simulator for 2 x 10(6) cycles. Bovine calf serum was used as lubricant, and the samples were weighed at regular intervals during the test. The predictive role of the lambda ratio on the wear behaviour was investigated. Three different configurations were tested to explore the wear rate for a broad range of lambda ratios. The results of these studies clearly showed that the femoral heads of 36 mm diameter had the best wear behaviour with respect to the other two smaller configurations tested. From a predictive point of view, the lambda ratios associated with the configurations tested could clearly indicate that the femoral heads of 36 mm diameter worked in the mixed-lubrication regime (lambda > 1); all the smallest configurations (28mm size) had lambda < 1, thus showing their aptitude to work in the boundary lubrication regime, with substantially higher volume depletion due to wear. The lambda values associated with the 32 mm size varied in a range around 1 (0.95 < lambda < 1.16), suggesting the possibility of operating in the mixed-lubrication regime.     

Optimization of process parameters for recycling of mill scale using Taguchi experimenal design

The present work submits an investigation about the optimum process parameters and quality improvement of mill scale recycling. With increasing concerns on environmental issues, the recycling of materials of all types has become an important issue. In this paper, an optimization method is developed to improve quality in mill scale recycling. The optimum configuration of process parameters to achieve high metallization efficiency was determined by experiments. The Taguchi method, the signal-to-noise (S/N) ratio, the analysis of variance (ANOVA) and rsponse surface optimization are employed to find the main effects and to determine their optimum process parameters. The significant process parameters were identified and their effects on mill scale recycling were studied. Finally, a confirmation experiment with the optimal levels of the process parameters was carried out to demonstrate the effectiveness of the Taguchi method.     

A study of the wear of some materials in connection with total hip replacement

Failure of total hip prostheses due to wear is examined. It is concluded that wearing out of these devices should not be a problem. However, it is desirable to look for materials of improved wear resistance due to possibilities of long-term response to wear debris.A series of experiments is described to evaluate the wear resistance of candidate materials on an annular wear tester. The results indicate that the wear resistance of ultra high molecular weight polyethylene may be improved by increase in molecular weight, by irradiation or by the use of fillers. Pyrolytic carbon containing silicon is also a good candidate.     

The effect of socket design, materials and liner thickness on the wear of the porous coated anatomic total hip replacement.

The wear of joint replacement prostheses represents the greatest challenge to their continued development. Parameters such as polyethylene quality, liner thickness and metal backing have all been implicated as potential detractors in the search for the lowest-wearing socket. This study examined the effect of these parameters through an extensive study of the two versions of the porous coated anatomic (PCA) hip prosthesis (one-piece socket and snaplock socket). For the whole cohort the wear rate was found to be 88 (SE 10) mm3/year and the clinical wear factor was 2.00 (SE 0.28) x 10(-6) mm3/N m. When the two socket types were investigated individually, the wear factors found were 2.39 (SE 0.44) x 10(-6)mm3/N m and 0.99 (SE 0.25) x 10(-6) mm3/N m for the one-piece and snaplock, respectively. This illustrates that the metal backing per se does not predispose these sockets to rapid wear. The good wear performance of the snaplock liner may be attributed to the high quality of the ultra-high molecular weight polyethylene (UHMWPE) used and the shorter implantation period compared to that for the one-piece design. No correlation was found between the thickness of the liner and the clinical wear factor. Within the range of thicknesses tested here, UHMWPE thickness is not an influential parameter for the hip prosthesis and this is confirmed