Weng Tat Chan

NEURAL NETWORK: AN ALTERNATIVE TO PILE DRIVING FORMULAS

Abstract Artificial neural networks are capable of learning complex nonlinear relationships from a large amount of accumulated data, and similar to human brains, are noise and fault tolerant. This unique capacity suggests that neural networks would be very useful in certain geotechnical engineering applications. A back-propagation network is set up and trained to predict the pile bearing capacity from dynamic testing data. The trained network produces better results than a pile driving formula approach. The effects of various network parameters on the network results are examined in detail. The general understanding developed is potentially useful for the application of neural networks in other geotechnical engineering problems.

Backcalculation Analysis of Pavement-Layer Moduli Using Genetic Algorithms

Most existing iterative backcalculation programs for pavement layer moduli arrive at their solutions by minimizing an objective function related to the differences between computed and measured surface deflections. Unfortunately, the solution surface of the backcalculation problem of pavement-layer moduli is known to contain many local minima. A potentially good backcalculation procedure would be one that has a strong global search ability to overcome the problem of local minima. The genetic algorithm (GA) is a technique that satisfies this requirement. The development of a backcalculation program known as NUS-GABACK using the genetic-algorithm approach is presented, along with the formulation and operations of the program. A detailed performance evaluation of the GA-based method is made against four other programs by solving five backcalculation problems with different structural composition. It was found that NUS-GABACK performed comparably well against the other programs and demonstrated consistency in the accuracies of backcalculated moduli.

An application of genetic algorithms to precast production scheduling

Abstract There are two alternatives for production organisation in precast factories, namely the comprehensive method and the specialised method. Production scheduling under the specialised alternative has been found to be a difficult optimisation problem if heterogeneous elements are involved. A flow shop sequencing model is developed for this kind of production scheduling that considers the constraints encountered in actual practice. The model is optimised using a genetic algorithm (GA) approach. The results are compared with those obtained using classical heuristic rules in two examples that involve the objective of minimising the makespan or the total tardiness penalty. The comparison shows that the GA can obtain good schedules for the model, giving a family of solutions that are at least as good as those produced by the heuristic rules.

CONSTRAINT HANDLING METHODS IN PAVEMENT MAINTENANCE PROGRAMMING

The problem of pavement maintenance management at the network level is one of maintaining as high a level of serviceability as possible for a pavement network system through reactive and proactive repair actions, whilst optimising the use of available resources. This problem has traditionally been solved using techniques like mathematical programming and heuristic methods. Lately, the use of genetic algorithms (GAs) to solve resource allocation problems like the network pavement maintenance problem has received increased attention from researchers. GAs have been demonstrated to be better than traditional techniques in terms of solution quality and diversity. However, the performance of the GAs is affected by the method used to handle the many constraints present in the formulation of such resource allocation methods. Penalty as well as generate and repair methods are the usual techniques used to handle constraints, but these have their drawbacks in terms of computational efficiency and tendency to get trapped in sub-optimal solution spaces. The paper proposes a third method that is computationally more efficient than the previous methods. The method is based on prioritised allocation of resources to maintenance activities and the maximum utilisation of resources. Constraints on maximum resource availability are no longer used passively to check on solution feasibility (as in the previous methods) but are used to help generate feasible solutions during the resource allocation phase of the algorithm itself. It is demonstrated that the GA with the prioritised resource allocation method (PRAM) outperforms the traditional GA with repair or penalty methods. PRAM was able to consistently outperform the other two GA based methods, both in terms of solution quality as well as computational time. It is concluded that PRAM can be used as the basis of more efficient resource allocation procedures in the area of pavement maintenance management.

Collaborative scheduling over the internet

The modern global economy compels people to communicate, collaborate, and cooperate efficiently. To date, information technology has facilitated this communication and collaboration by making the transfer of information quicker and more efficient through the use of digital and computer communications. However, it is argued that the efficient transfer of information is not enough; people still must coordinate their decisions based on the information exchanged. There is always the danger that the volume of information and the pace at which the information is exchanged will overwhelm the people, and they will not realize the implications of their decisions based on this information. This is especially so when the information exchange is across organizational boundaries. Computer support for such decision coordination is still very much lacking. This article looks at the subject of decision coordination in a very task-specific area, that of precast fabrication scheduling. It discusses the issues involved and describes the proposed system model for implementing a groupware application for doing distributed scheduling. It also discusses the technology that will be used to implement the application.

Coordinated production scheduling of prefabricated building components

A decision support system (DSS) for coordinated prefabrication production scheduling is described in this paper. The proposed DSS supports the four key elements of production (re)scheduling, namely, conflict detection, determination of priority for conflict resolution, generation and evaluation of alternatives/options for conflict resolution, and ranking of outcomes for negotiation. The system combines the use of an explicit constraints-based scheduling model, genetic algorithms (GA) for the determination of scheduling parameters and conflict resolution priorities, and constraint programming (CP) to facilitate the rapid determination of several alternative schedules subject to the requirement of minimum disruption of existing plans.

Developing a Construction Safety Management System

Various safety management systems (SMS) have been introduced to the construction industry as a formal means of managing site safety. There are several shortcomings with the current perspective of SMS which is implemented among construction projects. Firstly, the focus is exclusively on contractors and their performance on safety related factors. However, safety is not only the contractors’ responsibility but should involve owners as well. Secondly, current SMS presents a fixed set of safety related factors and activities which does not account for the diversity of projects and organizations in construction industry. Lastly, current methods of SMS auditing take a check-and-verify approach and miss customization and development opportunities. This paper proposes an alternative view of SMS based on the well-established Capability Maturity Model Integration (CMMI) framework. The proposal includes owners as active participants/stakeholders during the development of SMS, and allows flexibility during the implementation of SMS. The framework encourages a shift perspective from SMS factor and activity auditing to system development in terms of capability and maturity. This promotes a long term view of SMS development beyond a single project and fosters development of safety management capability within the construction industry.

Benefits of Information Integration in Budget Planning for Pavement Management

The budget-allocation process in pavement management occurs at several levels of decision making. In normal practice, the central level is concerned with the allocation of the budget to regional agencies to improve the health of the overall road network, while the regions are responsible for the programming and scheduling of road maintenance in their respective networks. The information processed at different management levels differs, and information at one level may not be readily available to the other. As a result, the maintenance funds allocated by central management to regional highway agencies may not be best used. A budgetallocation procedure based on artificial-intelligence agents is presented; it takes into account information integration between the central and regional management levels by using multiagent systems. Agents are used to facilitate the interaction between decision makers at the two management levels and thus provide a means for vertical information integration. A hypothetical problem containing a two-level funds-allocation structure consisting of a central-level management and three regional highway agencies is investigated. Each regional agency has different road network profiles with different local-level objectives. A genetic algorithm is used as the optimization tool in the decision-making process carried out by each agent. The results are compared with conventional allocation approaches and a two-stage optimization approach. It is shown that, with better integration of information at the different management levels, the resulting budget-allocation strategy produces greater maintenance-cost savings for a given target level of pavement performance.

Pavement-Distress Data Collection System Based on Mobile Geographic Information System

In recent years, the development of geographical information systems (GIS) has shifted significantly from desktop applications to mobile field applications. Integrating the precision location data collection capability of a Global Positioning System (GPS) and the spatial processing power of mobile GIS provides an adequate system for field workers to collect data with increased efficiency and ease. This research aims to design and implement a pavement-distress data collection prototype for inspecting airport pavement condition, with the aid of mobile GIS. Unlike the traditional paper forms, the data collection system seeks to offer a customized user interface for distress data entry and a spatial query service. Initial preparation of the airport map layers was carried out on desktop computer using a variety of off-the-shelf software. The required files were then transferred to the personal digital assistant (PDA) for field trials. Two other handheld devices, namely the GPS receiver and digital camera, are attached to the PDA to capture precise location coordinates and digital photos, respectively, of the identified distress. The users are also able to obtain comprehensive search results of the distresses, in both an attribute table and map visualization form. To increase the search efficiency, an indexing method was implemented for the data collected. Trial experimental results show that the applied indexing method has significantly improved the search efficiency over the conventional exhaustive search method.

Solving the FS-RCPSP with hyper-heuristics: A policy-driven approach

Abstract In this paper, a problem in the area of scheduling, namely Fuzzy Stochastic Resource-Constrained Project Scheduling Problem (FS-RCPSP), is addressed. Like the original Resource-Constrained Project Scheduling Problem (RCPSP), the objective is to minimise the expected makespan of the project subject to precedence and resource constraints. However, due to mixed uncertainty comprising fuzziness and randomness in the estimates of activity durations, the makespan is a fuzzy stochastic number. Recognising both fuzziness and randomness in activity durations results in more robust schedules but the scheduling problem is harder to solve. A hyper-heuristic, named Self-adaptive Differential Evolution to Scheduling Policy (SADESP) is proposed to address this issue. SADESP has two key modules: (1) a module (policyEvolver) which evolves scheduling policy and (2) a dynamic scheduling procedure (dScheduler) which makes scheduling decisions using a particular scheduling policy. The performance of SADESP is benchmarked against CPLEX across an extensive set of 960 problems created with ProGen – a standardised problem generator for creating benchmark problems in scheduling. The results returned by SADESP for FS-RCPSP are very encouraging, both in terms of accuracy and computational performance.