This section first explains the organization of the dissertation and also the research methodology behind it, then clarifies the research goal in each topic.
The author was a member of the research group of Department of Mate- rial Handling and Logistic Systems of the Budapest University of Technology and Economics and worked on the research project ’Investigating construc- tion processes to improve logistics and information technology’ (project ID:
KTIA AIK 12-1-2013-0009) funded by the Hungarian Research and Technol- ogy Innovation Fund. Researches of the author about construction processes (started with the mentioned project) are from the equipment (vehicle/mobile machine) perspective.
From this point of view, the first step is the equipment selection. This topic is covered by the article [21], published during the project as a com- mon work of the author and other co-researchers. If the equipment is in our possession it is necessary to provide an optimal layout for the construction in order to minimize the movement of the equipment and the hazard of the construction site during the movement. Based on the previous knowledge of decision support and literature research (detailed in Chapter2) the author of the present dissertation with co-researcher realizes the following: the cri- teria of determining an optimal layout is missing and also that there is no right tool for decision making. So the Chapter 2 is about the chosen crite- ria and developing an appropriate MCDM by mixing the advantage of KIPA (Kindler and Papp) method and fuzzy logic.
Construction machinery can move through the construction site in op- eration or between operations. After having the optimal layout, before our equipment can start to move some conditions have to met. We have to pos- sess the tools to track, monitor the equipment. As there was no instruction about the right tools for this job, the second goal was the analysis of the pos- sibilities and making some suggestion about it, this is the topic of Chapter 3. It was important to examine roads too, besides construction site, because construction equipment and transportation vehicles move there as well. The next step is the path planning for the mobile machines/vehicles. This is not
1.3. Research Goals 5 detailed in this dissertation, however the author helped to develop a path planning algorithm especially for construction vehicles (considering the spe- cialties of construction sites) [22] (and also developing a travel planner [62]).
When the equipment is moving, their environment must be explored in order to ensure safety, help the operator, or in case of autonomous vehi- cles/machines help the work, navigation by landmarks. The author recog- nized that there are many algorithm dealing with obstacle recognition, how- ever there are no one, which can deal with partial 3D objects, despite the fact that in real scenarios these cases are much more frequent. So in Chapter4the author presents a pipeline which is capable of partial object recognition. This system is tested on an Automated Guided Vehicle (AGV). Unfortunately, it is slow for a high speed autonomous vehicle, but can be appropriate for a mobile machine. In Chapter5 another approach is presented which can be suitable for even less data than the method shown in Chapter4and even for autonomous driving. These methods are tested on urban objects, because the lack of construction site database for appropriate testing purposes. The opti- mization problems related to the mobile machine automation and the present dissertation is illustrated in Fig.1.1and summarized in Table1.1.
FIGURE1.1: Investigated topics of mobile machine automa- tion (green - in the dissertation, blue - related publication, red -
applications)
TABLE1.1: Mobile machine related problems investigated in the dissertation
Topic Problem Goal Proposed
solution Contribution
Map
Layout se- lection was based on subjective opinions
Finding op- timal layout for gener- ating opti- mal paths for mobile machines
Decision making method ca- pable of de- creasing un- certainty and subjectivity
Method de- velopment
Tracking
There was no decision support in this field
Selecting appropri- ate tools for tracking in case of a given task
Decision support cri- teria system
Developing the criteria system
Environment perception
The devices are equipped (on AGVs), but there was no ap- plicable method for partial recog- nition
Recognize objects for many (e.g.
navigation) purposes
Pipeline for partial object recognition
Local scale, pattern def- initions and method de- velopment
Environment perception
Far objects are ignored in case of au- tonomous driving (de- spite of the need)
Increase the LIDAR sen- sors’ detec- tion range and so the permissible speed of the vehicles as well
Applicable, simple and fast clas- sification method
Method de- velopment
7
Chapter 2
Hybrid MCDM in Construction Logistics
In case of transportation the available road network is a primary factor in- fluencing the routes, travel times, emission, etc. even the perceived envi- ronment (so safety). All these parameters are just as important in case of construction machines as in case of transportation vehicles and can be in- terpreted as costs. Construction processes cannot be optimized in point of view of machines (neither from point of view of supply, schedule, etc.) with- out an optimal layout defining the ’road network’ (traversable areas) of the construction. Generating feasible layouts or even Pareto front of them (op- timized with respect to some of the parameters) is achievable [33], [8]. The challenging task is to make the final decision and find the one, winner layout, because experts do not reach consensus on criteria, weights and compliance based on their subjective impressions. In this Chapter an appropriate tool is proposed for this decision making problem . The criteria of construction site layout planning (CSLP) and the above mentioned automated method was introduced in [171] and [196] and in [197] applied expected value calculation of hesitant fuzzy sets are compared to other defuzzification methods in order to prove its efficiency. The criteria development is the contribution of the co- author of the articles, the remaining results, formulated in the thesis (method development) are my contribution.
Construction site layout planning has been recognized as a critical step in construction planning by practitioners and researchers. Most construction resources require space on site. This is the case for materials, equipment, support facilities (e.g., trailers or parking lots), and demarcated areas (e.g., laydown areas, roads, or work space), but also for obstacles (e.g., trees or ex- isting buildings). Layout planning is to allocate site space to resources so that they can be accessible and functional during construction. CSLP involves identifying, sizing, and on-site-positioning of temporary facilities which may
include security fences, access roads, storage sheds, field offices, fabrication shops, sanitary facilities, electric power service, stockpiles of excavation, and batch plants [4]. It is recognized that a good layout has a significant impact on cost, timeliness, operational efficiency and the quality of construction, which manifests on the larger and more remote projects. Improper layout can result in a loss of productivity due to excessive travel time for laborers and equip- ment, or inefficiencies due to safety concerns. An important aim of the layout planning is to ensure continuous flow, which is one of the five principles of lean production. Reduction of bottlenecks, flexibility handling, and produc- tion, since in construction sites there are usually a lot of changeovers. In spite of its potential consequences, construction site layout generally receives lit- tle advanced planning, and hardly any planning during construction. It is often determined in an ad hoc manner at the time the siting requirement arises. Therefore, an effective construction site layout planning (CSLP) is ut- most importance for the success of a construction project. CSLP problems can be broadly divided into static and dynamic ones [140]. Creating layouts that change over time as construction progresses is termed dynamic layout planning. Dynamic layout planning enhances the efficiency of construction operations.
2.1 Criteria in Construction Site Layout Planning
The first step is to define the criteria, which will be used in the chosen MCDM method. The number and the types of the criteria determine the applicable MCDM methods. From the mathematical point of view there are two main types of criteria: quantitative criteria and qualitative criteria. Quantitative methods consider the actual transportation cost per the amount of moved materials, measured by mass each two locations [99]. Qualitative methods, on the other hand, consider a subjective numerical proximity weight to ex- press the desirability of having any two facilities close to each other on the layout, or safety of the layout. Both methods are used in this study. Accord- ing to [140], [201] these criteria are defined by:
• C1: Safety/OSH/fire protection aspects: Safety of construction opera- tions is usually affected by many factors. A comprehensive literature review and several field studies were conducted in order to explore and identify relevant and important practical considerations that can enhance the safety of construction operations [9]. A fire protection spe- cialist should approve the layouts.
2.1. Criteria in Construction Site Layout Planning 9
• C2: Operating costs: Operating costs become an important aspect if you have a long term project, in this type of constructions operating costs are more important than set-up costs.
• C3: Installation costs (set-up cost): The investment which is needed to build up the given construction site layout. It includes all the cost. The initial set-up cost will affect the tender price and the long term running cost.
• C4: Traveled distance by human labor: The sum of the distances that the employees should take during the operations.
• C5: Traveled distance by machines: The cost of the predictable displace- ments of the material handling equipment (mixers, cranes, excavators etc.) during the construction.
• C6: Material handling performance: The flow of raw materials, WIP (Work in Progress) and finished products between locations in the con- struction site. Material flow can be measured by cost multiplied by distance (measuring by unit is not applicable because of bulk materi- als) like „kg∗m” or „t∗m” , or if we have diversified the transportation cost for some material, than we can measure it by transportation cost.
• C7: Possibility of upgrading: Usually during the project there are de- lays, therefore we need more capacities to keep the deadline, in this situation upgrading possibilities are important.
• C8: Transport connections: The proximity of highways, roads, streets, and overpasses is an important aspect when you choose the optimal layout. Transportation ways and connections can determine the bottle- necks of a system [201].
• C9: Machine/workspace availability: It shows how easily machines can be moved out, or maintained.
• C10: Amount of onsite stackable supply: If there are huge supply dis- tances in the project, storage capacity can become very important.
We defined ten criteria, six quantitative and four qualitative, therefore we need a MCDM method which can manage both types of criteria.