The primary aim of structural optimization is to determine most suitable combination of design variables, so as to achieve satisfactory performance of structures subjected to behavior and geometric constraints imposed with goal of optimality being defined by conditions and three basic features of structural optimization problem are:
The design variables, objective function, constraints.

Over years, several techniques have been developed for solution of non linear programming problems and some of prominent techniques are: method of feasible directions, sequential unconstrained minimization technique, sequential linear programming and dynamic programming.

Stresses are developed at top and bottom fibres of critical section at stages of transfer of prestress and under service loads se conditions yield four inequalities expressed as:
(p/a ?p_e/z_t + M_g/z_l) > f_u, (p/a+p_e/z_b-M_g/z_b) < f_ct
(p/a-p_e/z_t+m_g+m_q/z_t) < f_cw, (p/a+p_e/z_b-m_g+m_q/z_b) > f_tw.

The structural optimization problem is generally expressed in mamatical form, involving design variables, objective function and constraints accordingly problem is expressed as:
Minimize Z = F(x), subject to G_j(x) < 0, j = 1,2?.m
Where x is design variable, represented by column vector of dimensions n and each design represented by x is a point in hyper-space defined by design variables, F(x) = objective or merit function, G_j(x) = constraints, m = number of constraints.

In structural designs, behavior constraints are usually imposed on stresses and displacements constraints prescribe global rigidity of structure side constraints are specified limitations (minimum or maximum) imposed on a design variable are usually explicit in form.

Bond has reported optimization studies on partially prestressed in which several parameters, such as span, volume of prestressing wire and steel bar reinforcements, cube strength, thickness of slab and permissible tensile stress are examined in relation to total cost of slab per unit area.

In sequential linear programming, non-linear objective function and constrains are linearized in vicinity of starting point and a new design point is obtained by solving linear programming problem and sequence of line arising in neighborhood and solving by linear programming is continued from new point till optimum is reached.

In design of prestressed concrete structural element objective function may comprise cost of steel and concrete in member and a general guide for selecting an objective function is that design should be optimized with respect to most important design property that can be meaningfully quantified and which is not constrained in advance.

The third major approach of structural optimization is based on concept developed by Prager and Taylor and this approach derived from extremum principles of structural mechanics and after advent of finite element techniques, large scale computer programmes have been developed to demonstrate practical utility of optimality criterion approach to design of minimum weight structures.

In design of fully prestressed (class 2) members, all constraints as given in a < a_p, ?_min < ? < ?_max are not valid, however in this case of partially prestressed members where cracks of limited width of permissible under working loads an additional constraint to impose limitations on width of crack is required and this can be expressed as: w < w_p, w = actual widt, w_p = permissible crack width.