scholarly journals Using a Random Forest Model to Predict the Location of Potential Damage on Asphalt Pavement

2021 ◽  
Vol 11 (21) ◽  
pp. 10396
Author(s):  
Xiaogang Guo ◽  
Peiwen Hao

Potential damage, eventually demonstrated as moisture damage on inner and in-situ road structures, is the most complex problem to predict, which costs lots of money, time, and natural resources for maintenance and even leads to safety problems. Traditional linear regression analysis cannot fit well with this multi-factor task in such in-field circumstances. Random Forest (RF) is a progressive nonlinear algorithm, which can combine all relative factors to gain accurate prediction and good explanation. In this study, an RF model is constructed for the prediction of potential damage. In addition, relative variable importance is analyzed to obtain the correlations between factors and potential damage separately. The results show that, through the optimization, the model achieved a good average accuracy of 83.33%. Finally, the controlling method for moisture damage is provided by combining the traditional analysis method and the RF model. In a word, RF is a prospective method in predictions and data mining for highway engineering. Trained with effective data, it can be multifunctional and powerful to solve hard problems.

2021 ◽  
Vol 11 (4) ◽  
pp. 1378
Author(s):  
Seung Hyun Lee ◽  
Jaeho Son

It has been pointed out that the act of carrying a heavy object that exceeds a certain weight by a worker at a construction site is a major factor that puts physical burden on the worker’s musculoskeletal system. However, due to the nature of the construction site, where there are a large number of workers simultaneously working in an irregular space, it is difficult to figure out the weight of the object carried by the worker in real time or keep track of the worker who carries the excess weight. This paper proposes a prototype system to track the weight of heavy objects carried by construction workers by developing smart safety shoes with FSR (Force Sensitive Resistor) sensors. The system consists of smart safety shoes with sensors attached, a mobile device for collecting initial sensing data, and a web-based server computer for storing, preprocessing and analyzing such data. The effectiveness and accuracy of the weight tracking system was verified through the experiments where a weight was lifted by each experimenter from +0 kg to +20 kg in 5 kg increments. The results of the experiment were analyzed by a newly developed machine learning based model, which adopts effective classification algorithms such as decision tree, random forest, gradient boosting algorithm (GBM), and light GBM. The average accuracy classifying the weight by each classification algorithm showed similar, but high accuracy in the following order: random forest (90.9%), light GBM (90.5%), decision tree (90.3%), and GBM (89%). Overall, the proposed weight tracking system has a significant 90.2% average accuracy in classifying how much weight each experimenter carries.


Biosensors ◽  
2021 ◽  
Vol 11 (2) ◽  
pp. 52
Author(s):  
Takehito Hananouchi ◽  
Yanjun Chen ◽  
Saeed Jerban ◽  
Masaru Teramoto ◽  
Yajun Ma ◽  
...  

In this study, we combined quantitative ultrashort echo time (UTE) magnetic resonance (MR) imaging and an investigation by a probing device with tri-axial force sensor to seek correlations with mechanical properties of human patellar cartilage for in situ evaluation of biomechanical properties. Cartilage blocks (15 × 20 × 5 mm3) were dissected from the patella of six donors; 5 mm square regions of interest from the cartilage blocks were imaged using UTE-MR imaging sequences (T2* and magnetization transfer ratio (MTR)), and mechanical properties were measured using a micro indentation device. Then, the vertical reaction force on the cartilage surface was measured while push-probing forward 3 mm with the probing device at a 30° tilt to the horizontal plane. The results showed a positive correlation between stiffness/elastic modulus and each predictor variable (UTE-T2* (r = 0.240 and 0.255, respectively, UTE-MTR (r = 0.378 and 0.379, respectively), and probing device force (r = 0.426 and 0.423, respectively). Furthermore, multiple linear regression analysis showed the combination of the three predictors had stronger correlation (adjusted r2 = 0.314 (stiffness), 0.323 (elastic), respectively). Our results demonstrate the potential for these non- and less- invasive methods for in situ evaluation of the mechanical properties of cartilage tissue.


2015 ◽  
Vol 15 (9) ◽  
pp. 5083-5097 ◽  
Author(s):  
M. D. Shaw ◽  
J. D. Lee ◽  
B. Davison ◽  
A. Vaughan ◽  
R. M. Purvis ◽  
...  

Abstract. Highly spatially resolved mixing ratios of benzene and toluene, nitrogen oxides (NOx) and ozone (O3) were measured in the atmospheric boundary layer above Greater London during the period 24 June to 9 July 2013 using a Dornier 228 aircraft. Toluene and benzene were determined in situ using a proton transfer reaction mass spectrometer (PTR-MS), NOx by dual-channel NOx chemiluminescence and O3 mixing ratios by UV absorption. Average mixing ratios observed over inner London at 360 ± 10 m a.g.l. were 0.20 ± 0.05, 0.28 ± 0.07, 13.2 ± 8.6, 21.0 ± 7.3 and 34.3 ± 15.2 ppbv for benzene, toluene, NO, NO2 and NOx respectively. Linear regression analysis between NO2, benzene and toluene mixing ratios yields a strong covariance, indicating that these compounds predominantly share the same or co-located sources within the city. Average mixing ratios measured at 360 ± 10 m a.g.l. over outer London were always lower than over inner London. Where traffic densities were highest, the toluene / benzene (T / B) concentration ratios were highest (average of 1.8 ± 0.5 ppbv ppbv-1), indicative of strong local sources. Daytime maxima in NOx, benzene and toluene mixing ratios were observed in the morning (~ 40 ppbv NOx, ~ 350 pptv toluene and ~ 200 pptv benzene) and in the mid-afternoon for ozone (~ 40 ppbv O3), all at 360 ± 10 m a.g.l.


INDIAN DRUGS ◽  
2016 ◽  
Vol 53 (01) ◽  
pp. 25-31
Author(s):  
M Priyanka ◽  
◽  
F. S. Dasankoppa ◽  
H. N Sholapur ◽  
NGN Swamy ◽  
...  

The poor bioavailability and the therapeutic effectiveness exhibited by the anti-depressant venlafaxine hydrochloride on oral administration is overcome by the use of ion-activated gel forming systems that are instilled as drops; these undergo gelation in the nasal cavity. The present study describes the design, characterization and evaluation of mucoadhesive nasal in situ gelling drug delivery of venlafaxine hydrochloride using different polymers like sodium alginate, HPMC and pectin in various concentrations. DSC studies revealed compatibility of the drug and excipients used. The in situ gels were characterized for physicochemical parameters, gelling ability, rheological studies, drug content, drug entrapment efficiency, in vitro mucoadhesive strength, water holding capacity, gel expansion coefficient and in vitro drug release studies. The amount of polymer blends was optimized using 23 full factorial design. The influence of experimental factors on percentage cumulative drug release at the end of 2 and 8 hours were investigated to get optimized formulation. The responses were analyzed using ANOVA and polynomial equation was generated for each response using multiple linear regression analysis. Optimized formulation, F9, containing 1.98% w/V sodium alginate, 0.64% w/V hydroxylpropyl methylcellulose, 0.99% w/V pectin showed percentage cumulative drug release of 19.33 and 80.44 at the end of 2 and 8 hours, respectively, which were close to the predicted values. The optimized formulation was subjected to stability study for three months at 300C /75% RH. The stability study revealed no significant change in pH, drug content and viscosity. Thus, venlafaxine hydrochloride nasal mucoadhesive in situ gel could be successfully formulated to improve bioavailability and to target the brain.


2017 ◽  
Author(s):  
Carlos J Corrada Bravo ◽  
Rafael Álvarez Berríos ◽  
T. Mitchell Aide

We developed a web-based cloud-hosted system that allow users to archive, listen, visualize, and annotate recordings. The system also provides tools to convert these annotations into datasets that can be used to train a computer to detect the presence or absence of a species. The algorithm used by the system was selected after comparing the accuracy and efficiency of three variants of a template-based classification. The algorithm computes a similarity vector by comparing a template of a species call with time increments across the spectrogram. Statistical features are extracted from this vector and used as input for a Random Forest classifier that predicts presence or absence of the species in the recording. The fastest algorithm variant had the highest average accuracy and specificity; therefore, it was implemented in the ARBIMON web-based system.


2021 ◽  
Author(s):  
Stefan Baltruschat ◽  
Steffen Bender ◽  
Jens Hartmann ◽  
Annika Nolte

<p>Water-rock-interactions in the saturated and unsaturated zone govern the natural variability of CO<sub>2</sub> in groundwater. However, anthropogenic pollutions such as excessive input of organic and inorganic fertilizers or sewage leakage can cause shifts in the carbonate-pH system in an aquifer. Additional dissolution of minerals and associated mobilization of harmful heavy metals are possible consequences. Anthropogenic groundwater pollution is especially an issue where a protective confining layer is absent. On the other hand, addressing an environmental hazard such as fertilizer input to a single parameter remain intricate due to the high number of possible competing reactions such as microbial-controlled redox reactions. To overcome these obstacles, machine learning based statistical methods become increasingly important.</p><p>This study attempt to predict the CO<sub>2 </sub>concentration in groundwater from a multi-feature selection by using Random Forest. For this purpose, groundwater chemistry data (in situ measured bulk parameter, major ions, nutrients, trace elements and more) from more than 23000 wells and springs in Germany were collected and homogenized in a single database. Measured or calculated CO<sub>2 </sub>concentrations<sub></sub>are used to train the Random Forest algorithm and later to validate model results. Beside chemistry data, features about hydrogeology, soil characteristics, land use land cover and climate factors serve as predictors to build the “forest”. The intention of this study is to establish comprehensive CO<sub>2 </sub>predictions based on surface and climate features and to identify trends in local CO<sub>2 </sub>production. Gained knowledge can be used as input for groundwater quality management processes and adaptation policies.</p>


2019 ◽  
Vol 262 ◽  
pp. 10013
Author(s):  
Beata Stankiewicz ◽  
Piotr Górski ◽  
Marcin Tatara

The dynamic behavior of lively footbridge is a complex problem. Recently there were numerous publications and recommendations related to the dynamic nature of footbridge. The complicated procedure which was set in a number of instructions and standards says nothing about actions aimed at avoiding critical frequency range in structure. In the paper, results of dynamic in-situ tests of cable-stayed all-GFRP (Glass Fiber Reinforced Polymer) footbridge are presented. Fiberline Footbridge, located in Kolding city in Denmark, was constructed in 1997 using 12 different pultruded profiles all made of GFRP material. The dynamic characteristics as well as vertical response of the tested footbridge under human excitation are given and discussed. Firstly, in order to estimate the dynamic properties of the footbridge, a series of free-decay responses under human jumping were conducted. The fundamental frequency of the analyzed structure was within a critical range. A methodology for footbridge classification with regard to their dynamic sensitivity was worked out and the correlation between the structure's properties and its dynamic response under pedestrian excitation was formulated. It was found that the analyzed footbridge fulfilled vibration comfort criteria elaborated by technical guide Sétra, however, more restricted acceleration limits suggested by Eurocode were not met.


2020 ◽  
Vol 37 (5) ◽  
pp. 267-277
Author(s):  
Maarten de Laat ◽  
Srecko Joksimovic ◽  
Dirk Ifenthaler

PurposeTo help workers make the right decision, over the years, technological solutions and workplace learning analytics systems have been designed to aid this process (Ruiz-Calleja et al., 2019). Recent developments in artificial intelligence (AI) have the potential to further revolutionise the integration of human and artificial learning and will impact human and machine collaboration during team work (Seeber et al., 2020).Design/methodology/approachComplex problem-solving has been identified as one of the key skills for the future workforce (Hager and Beckett, 2019). Problems faced by today's workforce emerge in situ and everyday workplace learning is seen as an effective way to develop the skills and experience workers need to embrace these problems (Campbell, 2005; Jonassen et al., 2006).FindingsIn this commentary the authors argue that the increased digitization of work and social interaction, combined with recent research on workplace learning analytics and AI opens up the possibility for designing automated real-time feedback systems capable of just-in-time, just-in-place support during complex problem-solving at work. As such, these systems can support augmented learning and professional development in situ.Originality/valueThe commentary reflects on the benefits of automated real-time feedback systems and argues for the need of shared research agenda to cohere research in the direction of AI-enabled workplace analytics and real-time feedback to support learning and development in the workplace.


1998 ◽  
Vol 274 (5) ◽  
pp. H1429-H1434 ◽  
Author(s):  
Takayuki Sato ◽  
Toshiaki Shishido ◽  
Toru Kawada ◽  
Hiroshi Miyano ◽  
Hiroshi Miyashita ◽  
...  

We developed a miniaturized conductance catheter for in situ rat left ventricular (LV) volumetry. After the validation study of the conductance volumetry in 11 rats, we characterized the end-systolic pressure-volume relationship (ESPVR) in 24 sinoaortic-denervated, vagotomized and urethan-anesthetized rats. Stroke volume (SV) measured with the conductance catheter correlated closely with that measured by electromagnetic flowmetry ( r > 0.95). No significant difference was found between the in situ LV end-diastolic volumes measured by conductance volumetry and postmortem morphometry; a linear regression analysis indicated that the correlation coefficient was 0.934, that the slope was not significantly different from 1, and that the intercept was not significantly different from 0. During cardiac sympathotonic conditions, the ESPVR was curvilinear. The estimated slope of ESPVR (end-systolic elastance, E es) by quadratic curve fitting at end-systolic pressure of 100 mmHg was 2,647 ± 846 mmHg/ml. Bilateral cervical and stellate ganglionectomy depressed contractility and made the ESPVR linear; a quadratic equation did not improve the fit. E es was 946 ± 55 mmHg/ml with the volume-axis ( V 0) intercept of 0.076 ± 0.007 ml. Administration of propranolol (1 mg/kg) further reduced E es (573 ± 61 mmHg/ml, P < 0.001) and increased V 0 slightly (0.091 ± 0.011 ml). We conclude that the conductance catheter method is useful for the assessment of the ESPVR of the in situ rat left ventricle and that the ESPVR displays contractility-dependent curvilinearity.


Sign in / Sign up

Export Citation Format

Share Document