DETECTING AND RESOLVING WORK-SPACE CONGESTIONS AND TIME-SPACE CONFLICT THROUGH 4D - MODELING IN THE MICRO LEVEL

(1)

DEGREE PROJECT,

REAL ESTATE AND CONSTRUCTION MANAGEMENT PROJECT MANAGEMENT

MASTER OF SCIENCE, 30 CREDITS, SECOND LEVEL STOCKHOLM, SWEDEN 2016

DETECTING AND RESOLVING WORK-SPACE

CONGESTIONS AND TIME-SPACE CONFLICT

THROUGH 4D - MODELING IN THE MICRO

LEVEL

ZENGSHITING ZHANG

TECHNOLOGY

(2)

Master of Science thesis

Title

Detecting and resolving work-space

conges-tions and time-space conflicts through 4D -

Modeling in the Micro level

Author

Zengshiting Zhang

Department

Master Thesis number

Architecture and the Built Environment

TRI-TA-FOB-PrK-MASTER-2016:1

Archive number

407 Supervisor

Väino Tarandi

Keywords

BIM, 4D, Construction, Work-space

conges-tion, Time space conflict, Location lag,

Flowline scheduling

Abstract

This degree project aims to find solutions to prevent construction process from delay

by detecting and resolving work-space congestions and time-space conflicts based

on 4D-modeling.

The purpose is to improve the work efficiency on the construction site of a hospital

project. Through a software experiment, the proposed solutions will be tested to see

if the conflicts on the construction site can be resolved or minimized. This is achieved

by following the construction phase of the NKS project from Skanska AB. The largest

hospital project in Sweden.

The theoretical framework focuses on the concepts of 4D, work-space congestions,

time-space conflicts, lean construction, last planner system, project organization as

well as reviewing a variety of literature regarding how to resolve the conflicts during

the construction process. The useful data and information have been gathered

through semi-structured interviews with project managers and workforce. The

obser-vations have been done on-site. Followed by 4D software experiment, by associating

the tasks to the different areas through LBS with the time constraint data, the

effi-ciency of work based on the quantity takeoff can be evaluated and thus it allows

(3)

pro-Acknowledgement

The thesis was carried out at the department of Real Estate and Construction

Man-agement at KTH, the Royal Institute of Technology. This report is the result of a

mas-ter thesis project carried out in conjunction with the Nya Karolinska Solna project

governed by Skanska AB.

I would first and foremost like to give thanks to the supervisor Göran Pettersson in

Skanska who offered me this great opportunity to carry out my study in NKS and

in-troduced me to many contacts. Also I'd like to thank professor Väino Tarandi at KTH

for the helpful guidance and continuous support throughout the whole project.

Additionally I want to express my appreciation to all the personnel involved in this

thesis project from NKS as well as to PhD Pouriya Parsanezhad from KTH who

pro-vided valuable knowledge and experience.

Finally, I would like to give a big thanks to Rob Krieger who provided me much help

in finalizing this thesis within a limited time frame.

(4)

Examensarbete

Titel

Detecting and resolving work-space

conges-tions and time-space conflicts through 4D -

Modeling in the Micro level

Författare

Zengshiting Zhang

Institution

Examensarbete Master nivå

Architecture and the Built Environment

TRI-TA-FOB-PrK-MASTER-2016:1

Arkiv nummer

407 Handledare

Väino Tarandi

Nyckelord

BIM, 4D, Construction, Work-space

conges-tion, Time space conflict, Location lag,

Flowline scheduling

Sammanfattning

Syftet med detta examensarbete är att finna lösningar för att förhindra

konstruktions-processen från förseningar genom att upptäcka och lösa arbetsplatsstörningar samt

tids- och utrymmeskonflikter baserat på 4D-modelering.

Målet är att effektivisera arbetet på byggarbetsplatsen för ett sjukhusprojekt. Genom

simuleringar kommer föreslagna lösningar, för att undvika konflikter på arbetsplatsen,

att prövas för att få bukt med eller minimera dessa konflikter. Detta uppnås genom att

följa byggnadsfasen av NKS-projektet från Skanska AB. Sveriges största

sjukhuspro-jekt.

Den teoretiska referensramen fokuserar på begreppen: 4D,

arbetsutrymmesbelast-ning, tids- och platskonflikt, lean construction, last planner-systemet,

projektorgani-sation samt granskning av litteratur som handlar om konfliktlösning under

byggpro-cessen. Användbara data och information har samlats in via semistrukturerade

inter-vjuer med projektledare och byggnadsarbetare. Observationer har gjorts på plats.

Följt av 4D-simulering, genom att sammankoppla uppgifter till de olika områdena

ge-nom LBS med tidsrestriktionsdata, kan effektiviteten av arbete som bygger på

mängdavtagning utvärderas. Det tillåter projektledare att utan svårighet förutse

po-tentiella konflikter. Så småningom kan tillämpningen av 4D-modellering hjälpa

plane-rare att visualisera ineffektiviteter i tidsplanen och därmed planera om

arbetsuppgif-terna innan de leder till förseningar.

(5)

1. Introduction

1.1 Background

Delay phenomena are considered as one of the biggest problems and challenges in the construction industry (

Owlabi and Amusan, 2014

). Many researchers and engineers have tried to solve the problem and optimize the results. This delay phenomena has many causes such as inadequate client’s finance and payment, improper planning and management, shortage in materials and problems with subcontractors (Alnuaimi and Mohsin, 2013).

This thesis mainly focuses on delays due to work-space conflicts in the construction process. More specifically, it addresses conflicts when there is congestion in the work-space, which is when they usually occur. This is when there are collisions of time schedule between different groups of people and when the work-space is occupied by materials and equipment. Conflicts occurring due to this haven’t been detailed enough and have not been predicted precisely in the long term. This is because spatial structure has not been studied sufficiently and because of inaccurate time scheduling and work plans.

With development of the construction industry, many new concepts and useful tools have been proposed in order to resolve the conflicts and to further improve the methods for reviewing these issues. Building Information Modeling (BIM) was the one concept that became more and more popular in both the design and the construction field during the past two decades. 3D, one of BIM’s techniques, is a basic tool to achieve visualization of projects especially for representing architecture in design. With the introduction of 4D techniques (3D plus timeframe), project activities can be better managed not only in the design aspect but also for the construction activities.

4D can reduce work accidents and improve work efficiency by providing visual activity information with a timeline schedule. According to the experiments from the previous projects (which will be illustrated in the literature review section later), 4D can easily detect the conflict between two activities regarding the time and space elements and thus enable project managers to reschedule activities accordingly, thus yielding better results.

(8)

1.2 Summary of previous studies

Wu and Chiu from Taiwan (2010) carried out research on developing conflict detection and analyzing modules based on a 4D CAD system. First, they categorized work-spaces into five aspects: building component spaces, site layout spaces, human work-spaces, equipment spaces and material spaces. Then they identified four major types of work-space conflicts, which are design conflict, safety hazard, damage conflict and congestion. Afterwards, they used Bentley MicroStation and integrated it with schedule information for the dynamic identification of space conflicts. Those conflicts can be detected based on work-spaces by using 4D Visualization. The detection and analysis system was created through Microsoft VB.Net which simultaneously detects and analyzes work-space conflicts with various working objects, as well as visually representing the location, size, scope and type of work-space conflicts (Wu and Chiu, 2010).

Another study was carried out by Akinci, Fischer, Kunz and Levitt (2000c). On the construction site, different types of spaces required different positional requirements. In order to automatically detect the conflicts by 4D SpaceGen, it is necessary to generate project-specific spaces required by activities and represent those spaces in four dimension (Akinci and Fischer, 2000b). They tried to clarify the different orientation and volumetric requirement descriptions in order to formalize mechanisms. Some past studies have generated four approaches: (1) Static or dynamic site layout planning (Eastman 1975; Tommelein and Zouein 1993; Choi and Flemming 1996; Alshawi 1997; Choo and Tommelein 1999). (2) Line of balance (O'Brien 1975; Birell 1981; Stradal and Cacha 1982; Halpin and Riggs 1992; Howell et al. 1993; Yamamoto and Wada 1993). (3) Path-planning (Latombe 1988; Morad et al. 1992). (4) Space-scheduling (Zouein and Tommelein 1993; Riley 1994; Thabet and Beliveau 1994; Choo and Tommelein1999). However, those studies did not describe the methods with enough detail. Thus, Akinci, Fischer, Kunz and Levitt (2000c) took a closer look at micro-level construction spaces and proposed two ways to solve spatial conflicts：using a transformation matrix to present the relationship between two graphical objects quantitatively and using the space – loaded 4D production models to make more realistic 4D simulations.

(9)

In terms of using 4D BIM for work-space planning, several Korean researchers have done a comprehensively systematic summarization of previous studies. Besides that, they also came up with a new framework for work-space planning by dividing the attributes of work-space and conflict (point out). There are many techniques and tools that are being used now such as Gantt chart, network diagram, critical path method (CPM), and line of balance (LOB) (Choi et al., 2014). However, some of those techniques cannot present the spatial feature of each activity and some of them are unable to consider more possibilities of the type of work. In order to have a better understanding, Choi et al. (2014) came up with a diagram with all the methods based on related studies (Figure.1).

(10)

Through this diagram, comparisons within each category between different papers are simply addressed. Starting in 2002, Akinci et al.(2002a,b) developed a 4D model which can automatically generate the work space for each object and also suggested a method to categorize and prioritize work space conflicts. Dawood and Mallasi (2006) followed by identifying work space congestion status by viewing the Pattern Execution and Critical Analysis of Site-space Organization (PECASO) system. Chavada et al. (2002) suggested AABB intersection and congestion test (CgS) with geometrical parameters between two work-space models for checking work space conflicts. So based on previous findings, Chavada et al. (2012) achieved further improvements by differentiating the work space requirements and work space occupations, integrating activity execution plan and material management plans and formalizing the procedure for work-space problem resolutions (Choi et al., 2014).

(11)

1.3 Theoretical Grounds

BIM

Building Information Modeling (BIM) is a way of working which enables everyone to better understand and interact with buildings by using digital models. With BIM, all team members should be working with the same standards, for instance, using the same format of model among different parties. Doing so, it can improve their actions and create more value from the combined efforts of people, process and technology, finally resulting in a greater whole life value for the asset. The whole life value is reflected in the many stages of the asset which starts from design, analysis, documentation, and fabrication, and then is followed by construction, logistics, operation, maintenance and finally results in renovation (Eastman et al., 2011).

3D

A 3D model contains geometric information like length, width and height of the building components. It makes the model more realistic compared to a 2D model. It also makes it easier to detect errors in the project. Furthermore, the client who is not an expert in construction would be better able to comprehend the product in 3D rather than in 2D (Velasco, 2013).

4D

A 4D model contain objects and assemblies that have schedule and time constraint data added to them. The information can be contained in the BIM or can be linked or otherwise associated (integrated and/or interoperable) with project design, construction activity scheduling, time sensitivity estimating and analysis systems. (Conover and Barnaby, 2009)

Work-space congestion

Work-space congestion is a situation that occurs when the work-space available for the resources of an activity or group of activities is either limited or smaller than the required work-space for such resources. (Chavada et al., 2012)

(12)

Time-space conflict

Time space conflict means that one activity’s space requirements interfere with another activity’s space requirements, or with work-in-space (Akinci et al., 2000a). The consequence of it can be problems with productivity, safety and constructibility on site. In the end, those problems translate into delays in the onset of activities (Akinci et al., 2000a).

Project organization

The main focus of project organization is Planning, Control, and Evaluation (Packendorff, 1994). From a project management perspective, it is more about enhancing participation than controlling or inspiring creativity in the plan (Kanter, 1983). The key words of project management are learning, leadership, renewal and innovation.

One of the important techniques for defining the task and goals in a specific way within project organization is called Work Breakdown Structure (WBS). The aim of WBS is to identify the activities of different groups and labors through the optimal sequence during production planning (Nathan, 1991).

Several planning tools have been adapted for achieving WBS. The most widely used was the Gantt chart, developed by Henry Gantt in the 1910s. The Gantt chart describes the tasks and events clearly based on the sequence according to the time scale.

Moreover, another tool which is efficient in planning is called CPM (Critical Path Method). It is based on the assumption that the duration of activities can be estimated in advance. CPM can evaluate the cost based on the changes of activities. The mathematical problem of the CPM is to find the optimal ratio between fixed and variable costs given that stipulated time limits are not exceeded (Wiest and Levy, 1969)

(13)

estimated accurately. In PERT, the cost varies directly with time, therefore time is the controlling factor. The less time spent, the more money saved and vice versa.

Finally, Line of Balance (LOB), which is also known as Vertical Production Method, is an essential communication and productivity analysis tool for projects that have repetitive work areas (Line-of-Balance, 2010). By using LOB, project managers can easily see if the activities are in good balance, which accordingly lets project managers know whether or not the activities are lagging behind schedule. It also forecasts future performance by receiving timely information regarding trouble areas, and then implements appropriate corrective action.

(14)

Lean construction

Hafey (2014) has stated that lean is about eliminating waste and making processes more efficient, with the end goal of better serving the customer. There are five fundamental principles for lean thinking.

a) Specify value from the customer’s perspective b) Identify the value stream

c) Confirm the flow through the whole process

d) Speed up the process to make sure everything is on time e) Achieve perfection by improving the solutions

Ren (2012) emphasized that subcontractors and suppliers are playing an increasingly important role in project construction. Additionally, the contribution of subcontractors to the total construction process can account for as much as 90% of the total value of a construction project

The traditional production method is to build up procedures based on constructors’ needs and convenience, then push the work until the next stage in order to meet clients’ requirements. However, due to insufficient communication and orientation, logistics was hardly coupled with information delivery which led to many mistakes being made, and materials wasted during the whole process. In opposition to this traditional way, lean construction proposes a new method. First, constructors need to fully understand client needs at the very beginning. They should know exactly what the client expectations and worries are. Second, instead of planning work from the initial stage to the final stage, constructors should premise on the final product and then plan everything back from there to the initial stage. Lastly, by combining the lean construction approach with the last planner approach, logistics can be optimally bound up with information flows. This reduces unnecessary waiting time and improves working efficiency significantly.

(15)

The concept of lean construction is related to the application of lean thinking to the construction industry. It is about improved delivery of the finished construction project to meet client needs. ( Koskela, 2009)

Last planner system

The Last Planner (also referred to as the Last Planner System) is a production planning system designed to produce predictable work flow and rapid learning in programming, design, construction and commissioning of projects (Hussain et al.,2014).

(16)

1.4 Research purpose and questions

Purpose

The aim of this thesis is to present a method of shortening the construction process by detecting work-space congestions and time-space conflicts. Moreover, the project aims to implement the method in Nya Karolinska Solna (NKS) project and test if the conflicts on the construction site can be resolved or minimized.

Questions

In order to achieve the purpose, the main goal had to be identified in the first place. By using Backward Chaining Method, the main goal was broken down into smaller targets that need to be achieved.

As the figure. 2 shows below, the main goal is to avoid delays and to keep construction time under control so that work efficiency can be improved and costs can be significantly lowered. In order to achieve this goal, the conflicts, as the strongest negative influential factor in the construction process, need to be clearly defined, and then a method needs to be generated to minimize them. Since most of the ongoing conflicts arise due to overlapping location or timeline schedule changes, it is essential to assign tasks to the corresponding location with the right time schedule and to keep track of those tasks in order to avoid potential problems. Therefore, finding an effective way of assigning tasks with less mistakes and resolving them as soon as they arise is the optimal means to solving this critical problem in construction management.

(17)

Figure.2 Backwards chaining method

Research questions have been raised based on the research goal. The answers will be stated in the end of this thesis.

The research questions are the following:

1. How are conflicts detected through 4D models? 2. Can those conflicts be solved permanently?

3. How can timeline schedules be re-arranged in order to maximize performance?

4. What is the potential risk of re-arranging the timelines for the tasks? 5. Is it possible to prevent the construction process from having delays?

(18)

2. Methodology

This section explains the methods used in this thesis. A number of methods were used in conjunction with one another. First, in order to achieve the purpose of this study, an extensive literature review was done before and during the study. Second, interviews were conducted with many managers and workers on site. Through the interviews, data was collected effectively which helps the thesis to be more accurate. Third, an actual construction project plan was assessed an optimized through use of a software planner tool. The results of the tool were then assessed. The actual meaning and implementation of each method are further illustrated below.

2.1 The literature review

A literature review has three main purposes. First, to broaden the horizons of researchers. Second, to prevent the researchers from repeating work that has already been done. And third, to understand the strengths and weaknesses of the previous studies.

The literature review has been carried out to provide general background knowledge for this study. It mainly focused on 4D and BIM. The type of sources vary from scientific papers, reports and master thesis. Primary sources of information are based on scientific databases, KTH Primo and online researching.

From the literature review, the previous studies were divided into different categories according to the tools they used (4D, 4D CAD, 4D BIM). Through the comparison, gaps and limitations could be identified, and that led to new research questions (Silva, 2009).

(19)

2.2 Semi-structured interview

According to Berg (1998:59), interviews can be categorized into three groups: Structured, informal and semi-structured interview. In this thesis, semi-structure interviews were chosen as main method.

Bjørnholt and Farstad, (2012) further defined this method:

‘’A semi-structured interview is a method of research used in the social sciences. While a structured interview has a rigorous set of questions which does not allow one to divert, a semi-structured interview is open. This allows new ideas to be brought up during the interview. The interviewer in a semi-structured interview generally has a framework of themes to be explored.’’

The advantage of semi-structured interview is that the interviewers have enough time to prepare their questions and build up a good structure so that they can successfully lead the conversation during the interview. Another advantage is that the interviewees have more freedom to express their own view and perspective. Thus it can lead to more reliable, comparable qualitative data (Cohen and Crabtree, 2006)

The semi-structured interviews for this project were divided into two steps. The first step was done before starting to write the thesis (end of February to beginning of April, 2015) and consisted of designing and conducting one interview with each of seven mangers of the NKS project from different departments. The questions were designed to get to know some basic project information and to understand their opinions about combining current tasks with 4D BIM. The second step was to conduct further interviews with the managers from step one, if required based on research needs, as well as to conduct interviews with additional contributors to the NKS project including on-site area managers, workers, BIM specialist and other relevant staff. This allowed for the collection of useful information, the receipt of feedback, and for adjustment of research direction.

The general interview with both NKS staff and KTH staff is recorded below (Figure. 3) based on the timeframe. Regarding the detailed content of each interview, please refer to the appendix of this thesis.

(20)

(21)

2.3 Observation

Observation is important when analyzing construction project work. It is a way of gathering data by watching behavior and events, and noting physical characteristics in their natural setting. The main advantage of this method is that the researcher can focus on community and know the immediate impact of events. It allows the researcher to get an inside view of reality.

There are two types of observation, participant observation and non-participant observation. Non-participant observation is when the researcher attempts to observe people without letting them know. This means the researcher is not involved in the event. Participant observation is when the people who the researcher is observing and studying are cognizant of the researcher’s presence. It assumes that the researcher observes the events and relevant people closely. In some situations, observers may even need to live or work in that area (cite). For this study, non-participant observation is selected. This means that the author observed the time difference and relation between each construction activity in a natural setting on site without those observed being explicitly aware of the researcher’s presence.

(22)

The observation information is recorded below (Figure. 4) based on the timeframe. Regarding the detailed finding from each observation, please refer to the appendix of this thesis.

Figure. 4 The records of observation

2.4 Backward Chaining Method

In order to define the research problem properly, Backward Chaining Method, as one of many methods utilized in research analysis, was adapted for this study. Backward Chaining methodology is like working backwards from the main goal. In order to answer the final main question, several sub-questions must be answered so that they can provide findings and resources that leads to the answer of the main question. Backward Chaining Method, as a tool, is used in the beginning of this study to clarify the key point and draw the direction.

(23)

3. Theory

3.1 Work-space congestion

On a construction site, various activities are usually carried out by subcontractors in a limited area. Every subcontractor needs specific work-space for worker movement, material allocation and equipment movement. Congestion usually occurs when the job site is crowded, which means there is an interference among two or more activities. One of the most common phenomena is that there is overlap in the spaces where subcontractors need to do their tasks. (Guo, 2002

)

.

In order to improve the productivity of overall work flow, project managers have to create a detailed space management plan with all the subcontractors’ activities involved. It not only must include the space allocation they required for activities, but also the specific time frame of doing it. Otherwise, multiple activity crews may clash with each other or conflicts may arise between workers and machines. Sometimes even experienced project managers are unable to identify all the potential conflicts. When conflicts occur, the schedule is affected and usually is delayed (Guo, 2002). Many bad consequences can arise as a result, for instance, safety accidents, collisions, stricture, and falling objects (Moon et al., 2014). Therefore, it is essential to minimize the occurrence of conflicts.

If the work-space size or range, as a key factor, can be determined in advance, the work-space congestion can be identified in advance as well. By doing this, the productivity will be improved accordingly. However, before studying and identifying the specific problems in the project case, several categories of work-space congestion have to be analyzed.

(24)

To identify the work-space congestion, there are three categories:

A. Objects conflict with objects B. Activities conflict with objects C. Activities conflict with activities

Category A represents a situation where the area allocated for objects A interferes with the one allocated for objects B (Figure. 5). The objects could be arbitrary materials or equipment belonging to some activity. The objects A are properly placed in zones 1 and 2, but the ones in zone 4 occupy parts of the walking path and the area in zone 3. The dark blue area depicts the overlapping region and conflict area. This situation will affect construction process. For instance, suppose equipment for a ventilation installation was delivered to the site in zone 3 on Monday and it needs to be prepared at least one day before the installation process can be started. Additionally suppose that there were many side boards and boxes together with the materials placed in zone 4 overlapping with the corridor. If the phase do not clean up the area and move the materials away, moving in the ventilation equipment will generate obstacles. If the conflict can be identified in advance, the project manager can reschedule the tasks so that the problem will be solved earlier.

(25)

Category B represents a situation where the area allocated to an object interferes with the working area of an activity (figure 6). There are many objects stored in zones 1 and 2 while at the same time a group of people is doing their tasks in zones 1, 2 and 3. It is obvious that the work-spaces for both objects and activity overlap (dark blue area in the picture). This state will cause delay for the construction work. For example, suppose some equipment and materials are placed in zone 1. At the same time, several workers enter the space to do flooring according to the schedule. Without emptying the whole zone, the workers cannot start their job. As a result, they need to wait until all the objects have been moved away and then start flooring. In this way, the duration of flooring will be extended which may also affect the following activities. If the project manager can identify the conflict in advance, the tasks can be finished more effectively.

(26)

Category C represents a typical work-space congestion where two activity work schedules overlap, which means they occupy the same working space during the same time period (Figure. 7). As the picture shows below, activity B is taking place in zones 3 and 4 whereas activity A is taking place in zones 1, 2 and 3. Under this situation, zone 3 became the sharing area for both activities. Also the walking path near zone 3 is occupied by Activity A. So, suppose participants in activity A are doing pipework in zones 1, 2, and 3, while at the same time participants in activity B are doing installations in zone 3. The workers of activity B need to place the materials somewhere near the tasks, which require the same working space as activity A. Besides that, delivering materials and moving equipment makes the walking path very crowded and chaotic. All this has a strong negative effect on the performance of each activity. If those conflicts could have been foreseen in advance, either the work-space or the work schedule could have been rearranged, and the conflicts avoided or significantly reduced.

Figure. 7 Activities conflict with activities

Thus, solving work-space congestion is considered as a crucial factor for improving work effectiveness.

(27)

3.2 Time space conflict

Time-space conflict focuses on date and location collision. This differs from work-space congestion which is mainly focused on location collision. More specifically, time-space conflict describes situations in which two activities overlap in time and their work-spaces interfere during the time overlap (Moon et al., 2014). It happens when phase planning is not scheduled for each activity correctly. Because the same location can have many activities going on during the whole construction period and there are thousands of different locations with uncountable activities carried out by subcontractors, it is hard to accurately assign the activities to each location on 2D or 3D platforms.

In order to reduce the conflict rate, simulation of time schedule combined with 3D model allows construction planners and analysts to experiment with and evaluate different scenarios during the planning phase. Therefore, if some activities clash with others, they can be detected in advance.

(28)

Several scenarios are stated below:

A. At the initial stage, time space conflict can be caused because of the overlapped schedule. For example, the project manager assigns task A to location H between 9:00-16:30 on one day while there is another group of people supposed to do task B at the same location between 11:00-18:00, as it shows on Figure. 8. So the two activities overlap between 11:00-16:30 (the area marked as dark color).The consequence is either the project manager makes a plan B immediately to separate these two activities, or the two groups of people have to stop their job and then spend some time to solve the problem.

Figure. 8 The situation when overlapped schedules are assigned

B. Another situation is when a time-space conflict occurs because of a delay despite the fact that the project manager assigned the tasks correctly. For example, Group A and Group B are supposed to do their tasks, respectively, from July 10th -16th and 17th- 25th separately in the same area (Figure.9). However, because of some unexpected reason (like someone got sick) in Group A, they have to postpone the deadline. Group A will instead finish their job on the 18th which means there will be two days (17th and 18th) of overlap with Group B on the timeline schedule (the area marked as dark color). In this case, the project manager has to delay the time for Group B to start their job.

(29)

C. Revolving one initial conflict can also often lead to the creation of additional conflicts, and then the need to resolve those. This is because almost all the activities are connected with each other, so changes made to one activity and its corresponding location(s) may cause other changes to the adjacent activities. For example, as figure.10 shows below, location A is required to store some materials to support activity W conducted in location B at the same time. After Activity W is finished, location A is required for doing activity Y. Now say that there was a conflict with an Activity Z in another part of the construction project, and a decision was made that Activity Z was a priority over Activity W and thus Activity Z had to be completed in location B before Activity W was completed. As a result both activities were delayed, particularly Activity W. This would also then lead to location A having to store the materials for Activity W for a longer period of time, which would subsequently lead to Activity Y having to be delayed. It also could have delay ramifications for subsequent activities to be conducted in Location B. So we can see how one or multiple new conflicts may arise from “solving” an initial conflict. Therefore a more comprehensive and all-encompassing solution must be adopted to avoid or resolve not only the initial conflict, but all the subsequent conflicts that could arise out of attempting to correct the initial one. The traditional ways like 2D drawing-based planning cannot detect all the conflicts effectively, especially the potential conflicts which will only occur when some changes are made to the original plan.

(30)

However, some other factors should be also taken into account which may lead to work-space congestion or time space conflict. For instance, inadequate or inaccurate information from the clients will disrupt the construction flow significantly. This happens when the information is delayed, or when clients change their minds, so the information cannot be delivered to the construction site on time with the correct content. Under this situation, either project managers have to stop doing current activities and wait for the correct information, or continue to guide the work with information that is not up to date. However, either way, there is a high chance of conflict creation. Thus it is important to keep lean thinking principles throughout the whole process in order to minimize the deviations. More specifically, customer value and requirements have to be specified at the beginning to ensure that there is a continuous flow in the process. Having close communication with clients, managers as well as employees will avoid mistakes and defects. The flow will then be optimized.

(31)

4 Case study – Nya Karolinska Solna

4.1 Project information

Nya Karolinska Solna (NKS) is the project name for the state-of-the-art hospital currently under construction next to Karolinska University Hospital in Solna, Stockholm. It is one of the largest projects ever in Sweden which is made up of many sizable subprojects. There are seven phases to the project in total. NKS was started in June, 2010 and it will continue until the end of 2017. The total gross area of the new hospital facilities consist of approximately 330,000 square meters. The main constructor, Skanska, is a world leading specialist in this area (see Figure. 11).

(32)

As figure. 12 shows below, Phase 1 and 2 are the Technological Building and Garage Car Park, which were already finished in December 2012 and May 2014, respectively. Phase 5 - the biggest part and also the most important one - is the main body of the hospital which occupies approximately 125,000 sqm2. This is about 40 percent of the whole project. The total cost for Phase 5 is around 4 billion SEK (excluding design). Phase 5 contains three buildings: 1:20, 1:30 and 1:40. They share the common floors from floors 1-3. From floor 4 onwards, buildings are built up separately with some parts connected. Phase 5B is the second part of the hospital which will be handed over by October 2017. Phase 6 is a research laboratory which is expected to be finished in July 2017. The last, phase 7, is now under construction as well. It will be handed over as a patient hotel combined with a garage.

(33)

4.2 Conflict identification

So far, the project has been successfully running through all phases. This thesis focuses on the micro level looking at U130 (the building in the middle of figure.13) in Phase 5 as an observation case in order to have a closer look at the construction site regarding the three scenarios stated in Chapter 3, the practical theory section. This was done because, at the time of writing, all the interior construction tasks are being implemented. Around 25 subcontractors are involved for the whole construction period. It is a good time to observe how those activities interact with each other and how products and materials are allocated. Furthermore, U130, as the middle part of the whole building, plays an important role in terms of coordinating the different tasks between U120 and U140. Since many materials needed to be delivered from U120 to U130 or U140, the workers have to pass through half or the whole space of U130. In this case, congestion may happen.

Figure. 13 Phase 5 2D model

The observation has been done with the area manager who is responsible for floors 3-12 of U130.

(34)

4.2.1 Category A: Objects conflict with Objects

Figure.14 below shows conflicts between objects. There was a lot of ventilation equipment stored in this room together with a pusher, cart, gypsum boards and many other materials. The storage layout of storing these materials was disordered. The ventilation equipment, which stands in the middle of figure.14, blocked the throughflow of the room. If any other big loads of materials needed to be moved in, it would create obstacles. Or, if any equipment in this room needed to be moved away, the disordered placement would also generate conflict. The consequence would be that the workers would have to put other equipment and materials in another place temporarily before entering this area or they would need to move the ventilation equipment away in order to have enough space for the incoming objects.

(35)

4.2.2 Category B: Objects conflict with Activities

Figure.15 shows objects in conflict with activities. It was taken on floor 7, U130 at the end of the corridor while there were two to three subcontractors doing their jobs. In one of the corridors, the worker was cutting board toward the wall. However, a few meters away, there was a long stick lying above the gypsum boards which blocked half of the path transversely. It was quite dangerous if someone passed through this corridor without noticing. Besides, it might also have generated conflicts while the worker was working around it. On another side of the corridor, there was a blue cart placed in the wrong way which occupied quite a big area of the corridor. Combining with the long stick, it became an obstacle for people passing by.

(36)

4.2.3 Category C: Activities conflict with Activities

The last type of conflict is activities conflicting with activities figure. 16 below shows that the flooring on Floor 4 had already been laid. However, there was a room in the end of corridor whose floor had not been laid. This was due to the fact that the former working group did not finish their task while the flooring group started their job. It also shows clearly in figure. 17 that there were still many boards, ropes, pipes, etc. stored inside the room. About 60 percent of the rooms were occupied by those materials, thus the subcontractor had to stop his job before reaching the room. The rest of the area on floor 4 had its flooring laid. As the consequence, the later subcontractor had to wait for the former subcontractor to remove the materials and fix the holes on the floor first, then flooring could be continued in this area. However, because of the incomplete task left uncompleted, flooring could not finished on time. The area manager had to rearrange a schedule with workers.

(37)

(38)

5 Software experiment

4D BIM Scheduling

Currently, Skanska is using Primavera P6 and Powerproject together with Naviswork, Revit, MagiCad and other software as BIM tools for project management and scheduling. These softwares have been successfully used for generating 3D and 4D models. However the models could be improved by adopting Vico software which combines 3D model and timeline schedule to achieve 4D simulation. More specifically, Vico helps planners divide the site into manageable areas per trade (Vico Software, 2015b) through LBS (location Breakdown Structure). After the tasks or activities are assigned to the different areas, the efficiency of work based on the quantity takeoff and the number of people can be evaluated. By doing this, all the activities are connected with their locations in the Vico system and it can be used to track and to arrange work sequences accurately. Followed by flow line scheduling, once the starting dates and closing dates are added up, the duration of all activities can be estimated. Therefore, the project manager can easily detect the conflicts and reschedule the activities to optimize the efficiency of the project. By using flow line scheduling, the construction schedule can be optimized.

5.1 Software implementation

5.1.1 Problem Assessment

Before officially starting the modeling phase, format issues were identified as a potentially big problem. As this thesis mentioned before, Vico Software has chosen as the main tool for combining 3D modeling and timeline scheduling. Literally, it supports many publishers (i.e, Tekla, Revit, ArchiCAD, AutoCAD MEP, and AutoCAD Architecture) and importers (i.e, IFC, CAD-Duct and SketchUp). However, due to the license limitation, only IFC files can be imported into Vico without requesting additional licenses. In this case, all the 3D models

(39)

Several obstacles need to be overcome when transferring these 3D models to be used in Vico.

A. First, Skanska has never used IFC files before, and the models for NKS were generated in dwg, rvt, nwd, nwc and nwf by using Revit, MagiCad, Naviswork or others, not in IFC. Therefore, it was necessary to transfer those formats to IFC for Vico in order to continue with the experiment.

B. Second, some of those models could not be generated as proper IFC files through the transfer. To use an example that reflected a common problem, in one case the size of file was only 5KB after transferring which means it did not contain enough information about elements and thus did not become a usable IFC file.

C. Third, even for the files that did transfer successfully to an IFC file, some of them still did not show as a proper image in Vico. It may have been because the sizes of files (around 153MB) were too big so the computer’s hard drive could not handle such a heavy workload. Or it could have been that some of those files were MEP models and they were created by MagiCAD which certainly include semantics and properties, however, were not constructed in full accordance with the structural requirements of an interoperable BIM model.

5.1.2 Solutions

Four 3D models needed to be successfully transferred and assessed for the purposes of this experiment, and in the end this was achieved. Three were MEP files and one was an architecture file, all from U130. The was no initial problem with having the architecture file in the correct .ifc format. Regarding the MEP model files, the BIM coordinator in NKS provided the MEP models in .dwg format at the beginning. As mentioned in obstacle B above, the attempt to convert .dwg to .ifc, initially was unsuccessful. However upon consulting with Patrik Malarholm, the owner of Vico software, he suggested publishing the DWG files from Revit to Vico directly. This worked, and the files published in Vico successfully, showing all the needed information.

(40)

5.2 Location Based Structure (LBS)

After four 3D models were published together on Vico, a building with floors, façade, ceiling and pipes could be seen. However, U130 contains 12 floors in total, and it would be too difficult to simulate all the associated activities for all the floors given the limited timeframe and scope of this study. So, due to the complexity and the size of the project, floor 7 was singled out and chosen as the experimental focus.

The initial pre-step using Vico software as it pertains to this experiment was to define the different physical locations (to break down the overall location, floor 7, into sub-locations) and associate each location with the features of the model within the Locations Systems Module of the software. With respect to Vico, this is called establishing location systems. Location systems are alternative Location Breakdown Structures. Every activity type is going to have a different work sequence, so it is important to pre-define the location system within this particular software module at the beginning so it can separate project tasks into the correct locations later.

(41)

5.2.1 The Experiment with LBS Manager

Once the underlying location system had been set-up in the Location Systems Module through the initial pre-step, the first regular step (Figure.19 and Figure.20) can be carried out - to define and divide the location (floor 7) into several areas in the main interface of the software in accordance with the location system breakdown. Floor 7 was divided into zone A, zone B, and zone C. The reason for dividing the area into smaller zones was to optimize the performance of the work-space when it comes to the different tasks. Figure 19 and Figure 20 show the 2D floor plan and 3D floor plan, respectively, which is the model being depicted in geometric form after the cut height and view depth are defined manually and entered into the software interface.

(42)

Figure. 20 Breakdown location into three parts

The second step (Figure. 21) is to create tasks and define them in the task manager. Usually, the tasks (which are also referred to as onsite activities) would have already been sorted out in another software. Here, Powerproject software is used to plan all activities. So, the selected activities will be exported to a spreadsheet from Powerproject and then inserted into the Task Manager.

(43)

The third step (Figure. 22) is to export takeoff item and their associated quantities from the floor 7 model into the Cost Planner sheet. The typical purpose of using the Cost Planner sheet would be to define target costs for a 5D model, however with 4D models (which is what this experiment pertains to) it is not necessary to add unit cost, consumption, waste factors, etc. into the cost planner sheet. Instead, only quantities are needed. Thus, in the cost planner, each takeoff item is listed with its associated quantity.

Figure. 22 Takeoff items with associated quantities

The fourth step (Figure.23) is to associate the information from the Cost Planner sheet to the tasks. Both Task Manager and Cost Planner need to be included in the view. After activating the cost planner sheet, the selected items can be dragged and then link to the corresponding task manager on the left. After assigning all the takeoff items to the tasks, it allows planner to calculate task durations based on detailed location-based quantity data. (Vico Software, 2015a )

(44)

Figure. 23 Viewing Task Manager and Cost Planner

The fifth step (Figure.24) is to assign tasks to Location Systems. (It should be noted that this function is only for a situation in which the planner already knows all the locations and their corresponding tasks). So, all the tasks get assigned to a specific location system and only the information from that location system will be filtered into task’s duration calculations.

(45)

Figure.25 below shows a general process of establishing LBS and assigning tasks. The goal is to increase the work crew’s productivity rate by tracking their migration and reducing the frequency of them moving around inefficiently.

(46)

5.3 Flowing scheduling

After breaking down the project into smaller locations, now it’s time to use these to plan and analyze the project work as it flows through each location. (Kenley and Seppänen, 2006)

Flow line schedule consists of LBS in the vertical axis and a calendar in the upper horizontal axis. There are a series of sloping lines representing different tasks within the schedule view. Each task is made up of sub-components of materials needed to complete that task which have been assigned to the task manager, which means each task includes the quantity of components. As tasks flow through the project, each one of them corresponds to its associated location in the vertical column.

5.3.1 The experiment of flowing scheduling

Figure.26 shows the flow line view of each activity corresponding with its location and timeline schedule. There are 29 tasks in total beginning on Jan 21st, 2015 and ending on Oct 9th, 2015. Taking an example of the first task - completion of first side boarding - it started on Jan 21st at zone b, then moved to zone c on Feb 5th and ended up at zone a on Feb 19th. The rest of the task shows that there are some conflicts between different activities because some flow lines overlapped with others during the same time period and some flow lines shared the same location at certain points. However, the perfect situation should enable each task to not overlap with its predecessor or successor, so all the crews can flow through the site location continuously with a clear logic relation and achieve the optimal productivity rate.

(47)

(48)

In order to detect the conflict and de-risk the schedule, it is essential to define schedule logic in Schedule Planner through a Gantt Chart view or Network view. The dependency dialogue box is activated in Figure.27 below when the planner draw a dependency link to another task. The options include FF/SF/FS/SS/SS+FF, buffer delays, and location delays. In this example FS is the appropriate option to select. FS means ‘Finish to Start’. So in this case, task number 2 will start after task number 1 is finished. Following this rule, the remaining tasks can be defined individually according to their own dependency. Picture 13 below shows the final version of schedule logic relationship.

Picture 12

Figure.27 The Gantt Chart view

(49)

5.3.2 Solution A

To ensure high usage and productivity in each location, the ASAP method can be introduced. ASAP, As Soon As Possible, indicates that under the external logic relationship tasks can start as soon as the location becomes available. (Note that when the planner starts to create a task, it will always start as a Continuous task or Paced task since it is the base of LBS scheduling). Take an example - as Figure.29 shows below, the yellow area marked between task 1 and task 2 is free. After selecting ASAP, task 2 splits at each zone (Figure.30). Taking a closer look at the calendar, the end date of the task also moved ahead. Therefore, by using ASAP, the location can be optimized in its usage and the tasks can be

(50)

(51)

5.3.3 Solution B

Another way of compressing the tasks’ duration is to use location lag. Location lag means that we can create a mandatory dependency rule for adjacent locations such that the location below needs to be completed before the location above can start. For example, as the three graphs of figure. 31, figure. 32, and figure.33 show below, under the dependency mode, the planner grabs the lowest location of task Paint MF Ceilings in Sluss and then draws a dependency line to the second location of Room Sealing 3. Then, the dialogue pops out so the planner can define logic relationships. Here, the location delay equals 1 (Figure. 32). By creating a location lag equal to 1, the tasks are staggered from each other to occur sequentially by location (Figure. 33). If location lag equals 2, it means it allows two areas to be completed before the next one can start. Further, the impact of doing location lag is that reallocating time to the particular location can facilitate scheduling.

Comparing the finish time between figure. 31 and figure. 33, the task Paint MF Ceilings in Sluss ends up on 18th September which is 7 days earlier than before.

(52)

Figure. 32 The view of setting

(53)

By combining ASAP and location lag in conjunction with LBS and time line scheduling, project planner can clearly see the conflicts during the whole construction process and then modify the dependencies of each task. Comparing the original planning view (Figure. 34). with the current planning view (Figure. 35), it is obvious that those four main conflicts have improved. The new planning view shows the total time for finishing those tasks as less than it was in the original planning view.

Figure. 34 The original planning view

(54)

6. Analysis

This section is the assessment of Vico software with respect to strengths and limitations of it as a BIM tool. Additionally, the questions raised in the Research Purpose and Questions section are answered in this section based on the software experiment.

6.1 Strengths and Opportunities for Beneficial Implementation

Regarding strengths, first off Vico software offers a wide range of functions not only focused on 4D modeling but also on 5D modeling (4D plus cost estimating). In the current NKS project, planners use scheduling tools and cost planning tools separately. However, it is possible to combine those project components by using one BIM tool and then follow up with reporting and data mining.

Second, the flow line view in Vico software is more readable than comparable tools. Compared to the schedule generated by Powerproject, Vico software contains one more element which is LBS. This enables clients and workers to understand the information more easily through the flow line view. Additionally, the flow line view is more effective when detecting conflicts because it aims to eliminate overlapping tasks and plan continuous work without conflicts crossing. As seen in the figures from the software experiment in Section 5, it can compress activity duration in certain locations in certain situations without introducing risks. Furthermore, using the time and space buffer function can mitigate the project bottlenecks as well.

Lastly, as it is important for the project to be visualized during the construction process, Vico software can also make 4D simulations for the project (when done with cost planning and schedule planning). So on the one hand, planners and managers can foresee the whole construction process visually for themselves, therefore allowing for potential conflicts to be easily discovered and for an adjustment plan to be implemented immediately. And on the other hand, when these planners and managers present the project to their clients and other people who are not experts in the construction field, this 4D simulation allows them to understand the information quickly, even though they are not well versed in

(55)

6.2 Limitations

There are a few limitations when adopting Vico software for planning construction activities. First, Vico software has certain requirements for the format of a model. In the case of Skanska, they did not use IFC format very often beforehand, thus they could not import some of their 3D models into the Vico software. It will be hard for Skanska to continue to work with Vico if they are not able to import fully comprehensive and accurate information. Secondly, even though Vico software offers really advanced possibilities for model manipulation not only in scheduling but also in cost planning, the planners and project managers in NKS have been using their current software and BIM tools for a long time. It will be time consuming for them to learn all the functions of Vico software and transfer all the data and information to the format that Vico requires. Finally, Vico software seems to have less capacity to run a big model due to file size limitations.

6.3 The answers for the questions

These questions were first proposed in the Research Purpose and Questions section. The answers are given below:

1. How are conflicts detected through 4D models?

By building up an LBS system with all associated elements of tasks and inputting a time line schedule, the conflicts will show up from the schedule planner view automatically.

2. How can those conflicts be solved permanently?

Once the conflicts are detected by the planner, they can be highlighted and communicated to avoid the issue. But with changes in the schedule due to adjustments to prevent these initially-detected conflicts, other conflicts may arise, or other unforeseeable conflicts may arise during the actual construction process due to issues that have nothing to do with poor planning. There needs to be continual monitoring as construction progresses. So in that sense not all conflicts can be pre-solved “permanently.” Yet since planners can easily visualize the

(56)

conflicts with the 4D model, they will be aware and preplan effectively in order to avoid risks to the greatest extent possible. So in general, in order to avoid any conflicts occurring as best as possible, it is very important to logically plan the schedule in the beginning.

3. How can timeline schedules be re-arranged in order to maximize performance?

This thesis has introduced two main methods. One of these was ASAP, As Soon As Possible, which allows tasks to start immediately by splitting the task itself when the location is available. The other one was Location Lag, which creates mandatory dependency from the related tasks based on the locations. By doing this, the location can only be used when the previous task is finished. Thus, it reduces the risk of overlapping and keep the performance stable.

4. What is the potential risk of re-arranging the timelines for the tasks?

The potential risk is having other conflicts arise after resolving the current one.

5. Is it possible to prevent the construction process from having delays?

It is possible to prevent delays. However, it requires not just re-scheduling the plan, but also quick problem solving ability, effective communication flow and sufficient finance support. However, by using visualized scheduling, project managers can minimize conflicts and improve their working efficiency.

(57)

7. Conclusion

Foreseeing conflicts is very important for the whole construction process. If project managers and planners can detect inefficiencies in advance, it allows them to avoid having activities delayed and having to rework them. Applying 4D modeling and flow line theory allows planners to visualize those inefficiencies in the schedule so that they can re-schedule the tasks before they lead to delays or other problems.

Nevertheless, there are methods based on theories other than flow line theory that can also be incorporated in the construction planning and execution process to increase efficiency and improve results. The true core of achieving optimal results is rooted in sound communication. Whatever the changes that are made by project managers, it is always essential that they are discussed with the subcontractors and that the information is delivered to the workers successfully. Furthermore, whenever planners reschedule, it is essential that it’s done in conjunction with the client’s needs and that the implementation is transparent.

The NKS project followed the Lean Construction Method and Last Planner System quite well through all construction phases for the portion of the project that was studied in this thesis. It was also well organized by using current BIM tools to achieve high productivity, and thus provided an effective use case for Vico software. The use of Vico software for improving construction planning and implementation efficiency, as described in this thesis, would provide improved efficiency for the NKS project, as well as any other similar construction project. 4D modeling through Vico software gives project managers another means and method to better schedule construction tasks onsite.

(58)

8. References

Akinci, B., Fischer, M., Levitt, R., and Carlson, R., (2000a). Formalization and Automation of Time-space Conflict Analysis. STANFORD UNIVERSITY

Akinci, B., and Fischer, M., (2000b). ”4D WorkPlanner; A Prototype System for Automated Time-Space Conflicts Analysis.” Proceedings of the ICCCBE-VIII, Stanford, California, August 14-17, 2000,

ASCE, 740-751.

Akinci, B., Fischer, M., Kunz, J., and Levitt, R., (2000c). Automated Generation of Work Spaces Required by Construction Activities. STANFORD UNIVERSITY Alnuaimi, A. S., and Mohsin, M.A., (2013), Causes of Delay in Completion of Construction Projects in Oman, International Conference on Innovations in Engineering and Technology (ICIET2013) Dec.25-26, 2013 Bangkok (Thailand) Alshawi, M., (1997). SPACE: An Integrated Environment for the Construction Environment. Time Research Institute, Salford University, Salford.

Birell, G., (1981). Construction Planning - Beyond the Critical Path. Journal of the Construction Division, ASCE, 106(CO3), 389-407.

Bjørnholt, M.,and Farstad, G. R., (2012). ”Am I rambling?’ On the advantages of interviewing couples together” (PDF). Qualitative Research 14 (1): 319.

doi:10.1177/1468794112459671

Chavada, R., Dawood, N., Kassem, M., (2012) Construction workspace

management: the development and application of a novel planning approach and tool, Journal of Information Technology in

Construction (ITcon), Vol. 17, pg. 213 – 236

Choi, B., and Flemming, U., (1996). Adaptation of a Layout Design System to a New Domain: Construction Site Layouts. Computing in Civil Engineering, New York, 711-717.

Choi, B., Lee H., Park, M., Cho, Y. K., and Kim, H., (2014) ”Framework for Work-Space Planning Using Four- Dimensional BIM in Construction Projects.” Journal of Construction Engineering and Management September Volume 140.Issue 9: 2. DOI. Web

Choo, H. Y., and Tommelein, I. D., (1999). Space Scheduling Using Flow Analysis. IGLC- 7, Berkeley, CA, 299-311.