A detailed examination of the pipe failure records for the City of Winnipeg, Canada, was performed. The failures were examined by age, type of failure, and degree of spatial and temporal clustering for each of the four most common failure types: namely, joint, circular crack, longitudinal split, and hole. A detailed examination of joint failure showed universal joints in cast iron pipes to be the most common type of joint failure. It was found that failures in general were more common in pipes 11–30 years old. It was not clear, however, whether these high values were due to age alone or to the higher proportion of the pipe in the ground being around that age. The four major types of failure were also found to be highly clustered in both time and space, indicating that a previous failure increases the likelihood of future failures in its immediate vicinity.

Several types of low stiffness flexible pipe have been introduced in he United States in the past 10 years. Because of its extremely low pipe stiffness, very stiff embedment soil is required to limit pipe deflection to a reasonable alue. This stiff embedment often consists of mechanically compacted crushed rock laced in shallow lifts. Unless this pipe is backfilled under near laboratory conditions, nstallation and compaction forces result in “bulges” or “flat spots” in the nstalled pipe. Often, these shapes are such that the buckling load which the pipe an withstand is reduced to as little as 10% to 20% of the calculated buckling esistance of an elliptically deformed pipe. Six buckling cases, two of thermoplastic ipe, three involving thermoset plastic, and one of metal pipe, are discussed.

The design of a safety grating for a culvert inlet as investigated using a physical model. The objectives of such a grating are: (1) To prevent a person who falls into the drainageway from being carried into and through the culvert; and (2) to prevent the person from being pinned against the grating. The basic design of the grating, recommended by previous researchers, was tested using a 1:10 scale model. Testing was performed over a wide spectrum of flood flows, including flows that exceed the culvert capacity. The grating has only a minor effect on the head‐discharge relationship, causing a slight increase‐in the headwater required to pass given discharge. However, the grating may act as a debris control structure and vigilant maintenance may be required to prevent clogging of the culvert inlet. Neutrally buoyant objects were introduced to the flow to assess the performance of the grating. Because of the parabolic shape of the grating, an object carried to the culvert tends to be pushed up the grating and out of the flow.

A new approach to optimal design, maximum capacity, and the best layout for parking maneuvers, of a corner lot for parking spaces is introduced. Certain assumptions and practical design principles are used to derive the model, a system of nonlinear equations. The model is applied to a rectangular corner lot and solved by iteration using generated real data for combinations of compact and standard cars. The results show that in each case several layouts give the same maximum capacity. To select the maximum capacity corresponding to the best (safest) layout, an overall factor of easiness for parking maneuvers, called Ease, has been derived and used. The optimal solution of one case is than modified to fit other design specifications and plotted as an example of an actual parking layout for a corner lot. The model has several features: (1) Intuitively plausible principles of the initial design; (2) generality of equations to handle any car size, any combination of car sizes, and different angles in different regions; (3) it is easy to solve via applying data; and (4) it introduces Ease, an important factor to measure easiness to park, which gives the safest layout, etc. Further generalizations of the model are possible directions for future research.

Pumping prediction models are important in the design and economic analysis of rigid pavements. A review of several pumping models, as well as the one developed by the writers, is presented in this paper. Pumping modeling is a very complex problem which involves many factors: slab properties, subbase properties, pavement properties, environmental factors, and traffic. The ideal pumping model should include all these factors. The pumping model presented here is based on field data from different sources and amount of deformation energy in the pavement structure. The volume of pumped material expressed in terms of the normalized pumping index is calculated as a function of the deformation energy produced by traffic. This volume is then adjusted to take into account the following conditions: subbase type, drainage conditions, load transfer adequacy, subgrade conditions, and climate. The model can be used in the optimal design of rigid pavements by predicting the effect of different design alternatives on the development of pumping.

In accordance with the concept of pavement serviceability rating, busride panel rating should logically be employed for assessing the pavement serviceability of bus lanes and other road networks that are used predominantly by bus traffic. This paper describes a test program which was conducted to study the need and feasibility of using bus raters for serviceability rating of bus‐traffic‐dominated roads in Singapore. The test program was carefully designed so as to allow for a meaningful comparison of bus panel ratings with car panel ratings. Test results indicated that the two modes of ratings were compatible and equally sensitive in differentiating pavements with different serviceability. However, bus panel raters were found to be more intolerant of pavement roughness than car panel raters. The study suggests that bus‐ride panel is a feasible means of rating pavement serviceability, and that there exists a need for its use in rating highway pavements carrying considerable bus traffic.

The objective of this work is to present the current state of the art in modeling the dispersion of traffic platoons, to present a new mechanism suggested by current models, and to test that mechanism with detailed field data for at least one site. A field study using video data collection is described. Travel time distributions from the field study are compared with the same from the microscopic TEXAS Model for Intersection Traffic; statistical similarity is shown; and the use of this model to simulate data is validated for this site. A mechanism based on Diffusion Theory and suggested by the Recurrence Model is presented and applied using empirical support from field data. This mechanism is shown to provide accurate platoon dispersion modeling for this site. A comparison with the Diffusion Model shows nearly exactly similar results and a comparison with the Recurrence Model shows improved results. Suggestions are made for future work in this area.

Existing models for estimating left turn capacity assume an average opposing flow rate throughout the cycle. While this assumption may be valid under random arrivals, it is evident that the presence of platoons results in different arrival patterns on the opposing approach during the green and red phases respectively. Pilot data collected at one intersection in the City of Chicago showed that the observed left‐turn capacity is inversely proportional to the platoon ratio and flow rate on the opposing approach. This indicates that progression for through traffic may have an adverse effect on the opposing left‐turns.