Feasibility study tool for semi-rigid joints design of high-rise buildings steel structures

. There are many ways to consider the final cost of the high-rise building structures and to define, which of their different variations are the most effective from different points of view. The research of Jaakko Haapio is conducted in Tampere University of Technology, which aims to develop a method that allows determining the manufacturing and installation costs of steel structures already at the tender phase while taking into account their details. This paper is aimed to make the analysis of the Feature-Based Costing Method for skeletal steel structures proposed by Jaakko Haapio. The most appropriate ways to improve the tool and to implement it in the Russian circumstances for high-rise building design are derived. Presented tool can be useful not only for the designers but, also, for the steel structures manufacturing organizations, which can help to utilize BIM technologies in the organization process and controlling on the factory.


Introduction
One of the greatest accents in the steel structures researches is the optimization of the design model. It takes into account factors of economics, ecological influence on the environment and safety of the structure. Prediction of the structure's manufacturing costs considers as a complicated operation. The real price of the different design decisions of the building structure depends on its geometrical characteristics, material, manufacturing quality, etc. There are many ways to consider the final cost of the structures and to define, which of their different variations are the most effective from different points of view. The research of Jaakko Haapio is conducted in Tampere University of Technology, which aims to develop a method that allows determining the manufacturing and installation costs of steel structures already at the tender phase while taking into account their details. In order to take into consideration all the features of the joint fabrication process, the macros for the modelling program Tekla can be used at the tender phase. The program is based on the calculation methods developed by Jaakko Haapio in [1], which use time-based cost functions for all the process of steel structures manufacturing, transportation and installation. In the program, also, carbon dioxide emissions of the structure can be easily derived from the Building Information Model (BIM) of the steel structure.
In the work Jaakko Haapio the economic analysis tool for skeletal steel structures was studied in order to make analysis and to find the most appropriate ways to optimize the design and construction processes. As a result of analysis a method was presented that allows to determine the manufacturing, transportation and installation costs of steel structures already at the tender phase while taking into account their details. This is possible by analysis of the initial building models at the tender phase. The special macros of the modelling program Tekla allows to take into consideration all the features of the joint fabrication process. Preset parameters were studied in the Jaakko Haapio time-based cost functions of all steel structures manufacturing processes at the workshop and during installations are known. The development and testing of these cost functions in Russian circumstances is the subject of the research.

Feature-Based Costing Method and other cost models
At the preliminary design step the cost and emissions can be calculated by the method presented in [1]. It is based on the Feature-Based Costing Method, where feature is an attribute which affects the costs of the structure during the project.
It involves dividing the manufacturing process into cost centres of a specified floor area and height, equipment suitable for executing the required process (i.e. drilling a hole) and a certain number of workers. These resources have a fixed per minute cost, whether the process is running or not. Some cost components are related only to process time, i.e. electricity consumption of the equipment. They are called variable costs. The time required by the process is the sum of non-productive time, i.e. fixing the profile to the equipment, and process time, i.e. drilling. Total process cost is the sum of the fixed cost multiplied by total time plus the variable cost multiplied by process time. Some processes may also involve non-time related cost components. These are added to time-dependent costs.
The basis of the feature-based costing method was described in the Jaakko Haapio doctoral thesis. The material flow of a skeletal steel assembly shown in Figure 1. The workshop is divided into cost centres, where single work phases are performed. The time spent at each cost centre is converted to costs including equipment, wages, energy, rents etc.
Total cost considers as a sum of the cost derived for each cost center. The generic form of the cost for each cost centre is:

Feasibility study methods development in Russia
The issues of steel structures optimization process was considered in the researches of Ya.M. Lichtarnicov [2], I.S. Cholopov [3][4][5] and others.
The function of cost and labour intensity for the steel structures optimization process is the key function in the research of Ya.M. Lichtarnicov, that was published in 1979 y. From that time price levels, technological operations, labour intensity and other factors have changed a lot. So, today it is extremely important to find a way to deal with dynamically updated database for cost function input data. The research mostly deal with feasibility study of different design variants of space arrangements and only one chapter for the joints.
In the cost analysis method of Ya. M. Lichtarnikov the cost of steel structures estimates based on mass calculation G, that consists of two parts: G0 -mass of main parts, Gs -mass of secondary parts. Main parts mass can be determined by stress distribution and buckling resistance of the structure. Secondary parts mostly uses by any design consideration and in order to provide adequate behaviour of the structure. The mass of the secondary parts can be taken into account by construction coefficient of mass ψ. So, the formulae for structure mass:

G=Gs+G0=ψ⋅G0
(2) The labour intensity can be calculated by multiplying the mass and the coefficients that Ya. M. Lichtarnikov get from the analysis of big database of different structures costs. The coefficient of seriation that can be taken into account in order to provide better accuracy of manufacturing cost calculation.

Fig. 2.
The coefficient of seriation KC dependence from the quantity of equal parts in the one production run diagram [2] For each steel grade the reduction and correction coefficients should be included in analysis that can be taken from tables 1-2. [2]  For construction of main steel details from different steel grades the reduction coefficient α is used, that has the square root dependence for the labor intensity calculation (Table 2). The example of correlation between cost of steel structures transportation and the region of construction for columns and gantry girders can be determined from Table 3. [2]  While dealing with the installation cost it is necessary to include in analysis the steel grade and the location of the site. The correction coefficients for steel grade was presented in table 3. The location of construction site can be taken into considerations by the correction coefficient Kp from table 4. [2] The steel grade correction coefficient for different types of structures is given in table 5. [2] The correlation between the installation costs from the height and the cost of painting for different weights of structures is shown in the tables 5-6. [2]   There are more different correction coefficients and more detailed cost functions, that can be derived from the research of Ya. M. Lichtarnikov [2]. It may be included in analysis in the future improvement of the method, at least, as coefficient that provide relative functions of different feature dependences, but in that thesis was excluded from calculations.

Semi-rigid joints definition by Eurocode 3
In Eurocode 3 [6] connections are classified regarding their strength and their stiffness. The stiffness classification is clear from Figure 1. It is shown (see Gomes et al. [7]) however that if plastic rotations are adopted the use of initial stiffness as a unique description parameter is not correct. In other words the initial stiffness of the connection is not enough to classify the connection properties. In this research that method of semi-rigid joints description is enough. Fig. 3. Classification of beam-to-column connections by stiffness according to the Eurocode 3 -Annex J (revised): a) unbraced frames and b) braced frames. The region studied in the paper is denoted by arrow.

Design example
The third numerical example is a three bay, ten storey steel frame designed by Xu and Grierson [8]. Kameshki and Saka [9] as well as Foley and Schinler [10] performed weight optimization on the same example using Genetic Algorithms and evolutionary computation. The frame configuration, dimensions and loading are shown in Fig. 4. The used steel grade is S235, with a modulus of elasticity of 210 000 MPa and yield stress of 235 MPa.

Design procedure description
Basic times of different manufacturing operations are evaluated according to a fabrication cost model of joints developed at the Laboratory of the Design Optimization and Environment Engineering (LOCIE) of Polytech'Savoie (France). This cost model was first developed by Hamchaoui [11] and incorporated into a computerized module for joint design. Bel Hadj Ali updated the cost model for structural optimization with semi-rigid joints [12,13].
Design is performed considering only members as design variables while beam-tocolumn connections are specified to be of three type. Column bases are supposed to be rigid. Optimization variables are thus limited to 10 groups of beam members and 10 groups of column members. Beam members at each storey level are to have the same European IPE section while exterior and interior column members are to have the same European HEB section over two stories (Fig. 4).

Results
Optimum designs obtained for frames with rigid and semi-rigid connections are presented in Tables 7, for unbraced and braced frame configurations, respectively.

Conclusions
Structural optimization has been widely studied over the last decades and extensive work has been done in the case of optimal design of steel frames [14][15][16][17][18][19][20]. However, engineers have few tools to approach cost optimization in a systematic manner.
Here are som developments that should be made: 1. Dynamic data collection. The cost of hot-rolled steel profiles varies significantly for fairly short periods of time, thus prices monitoring for products, wages, prices for consumables and electricity can be updated by the dynamically replenished database, which will allow to change the specific coefficients of costs.