Capacity-constrained traffic forecasting model

Projecting future traffic is an important stage in any traffic and transportation planning study. Accurate traffic forecasting is vital for transportation planning, highway safety evaluation, traffic operations analysis, and geometric and pavement design among others. In view of its importance, this paper introduces a regression-based traffic forecasting methodology for a one dimensional capacity-constrained highway. Five different prediction functions are tested; the best was selected according to the accuracy of projections against historical traffic data. The three-parameter logistic function produced more accurate projections compared to other functions tested when highway capacity constraints were taken into consideration. The R ² values at various test locations ranged from 88% to 98%, indicating good prediction capability. Using the Fisher's information matrix approach, the t-statistic test showed all parameters in the logistic function were highly statistically significant. To evaluate reliability of projections, predictive intervals were calculated at a 95% level of confidence. Predictions using the logistic function were also compared to those predicted using the compound growth rate and linear regression methods. The results show that the proposed methodology generates much more reasonable projections than current practices.     

Filling up gaps in wave data with genetic programming

A given time series of significant wave heights invariably contains smaller or larger gaps or missing values due to a variety of reasons ranging from instrument failures to loss of recorders following human interference. In-filling of missing information is widely reported and well documented for variables like rainfall and river flow, but not for the wave height observations made by rider buoys. This paper attempts to tackle this problem through one of the latest soft computing tools, namely, genetic programming (GP). The missing information in hourly significant wave height observations at one of the data buoy stations maintained by the US National Data Buoy Center is filled up by developing GP models through spatial correlations. The gap lengths of different orders are artificially created and filled up by appropriate GP programs. The results are also compared with those derived using artificial neural networks (ANN). In general, it is found that the in-filling done by GP rivals that by ANN and many times becomes more satisfactory, especially when the gap lengths are smaller. Although the accuracy involved reduces as the amount of gap increases, the missing values for a long duration of a month or so can be filled up with a maximum average error up to 0.21 m in the high seas.     

Derivation of wave spectrum using data driven methods

The current techniques of derivation of a wave spectrum from given values of design wave parameters, like significant wave height and average wave period, are fraught with considerable uncertainties. This leaves scope for alternative approaches. The reported work proposes potential applications of two recent data driven methods, namely support vector regression (SVR) and model tree (MT), to obtain the wave spectra. In the present study the above tools were used to estimate wave spectra at two locations: no. 44008 maintained by National Data Buoy Centre (NDBC) in the Gulf of Maine, USA and ‘DS5’ monitored by National Institute of Ocean Technology (NIOT) in Bay of Bengal, India. The choice of these two locations facilitated the comparison of model performances in different geographical areas. The SVR and MT models were developed in order to estimate the wave surface spectral density over a wide range of wave frequencies out of average wave parameters of significant wave height and average zero-cross wave period. The models were trained and tested using randomly selected sea states. Both MT and SVR were able to derive the spectral shapes satisfactorily as reflected in high values of the correlation coefficients and low values of root mean square error and mean square error.