Some possibilities of optimization in the Romanian firm management.

Cociu, Nicolae ; Ignaton, Elemer ; Taroata, Anghel 等

1. INTRODUCTION

Studying the firm management optimization aims to increase its competitiveness and to ensure a rational decisional and managerial process.

Choosing mathematical models and methods is made according to the actual conditions of the Romanian firm, the information that can be collected and processed in due time, its costs, the competition strengths, in order to obtain the competition advantage.

Using the mathematical models, the optimization methods and the information technology in management and the firm's activity leads to an increase of its efficiency, a competitive advantage of low costs and better quality for the products placed on market.

The firm's management system is defined as an assembly of elements with decisional, managerial, informational and motivational characteristics; it triggers all management processes and relations in order to obtain a better efficiency and effectiveness. When projecting the firm's management system, a few elements should be considered: the firm's size and goals, its profile and complexity, the capital structure and dimension, the human resources, the firm's position in its competition with other firms, the conception of the owners and managers.

The management system can be structured as follows: the organizational, informational, decisional, management methods and techniques subsystems. (b. Cociu, 1999)

2. OPTIMIZATION POSSIBILITY

Optimization represents the choice and enforcement of the best solution of many possible ones. A model is a simplified or abstracted representation of reality. Often in practice it is difficult to simultaneously obtain simplified and representative characteristics of the studied reality.

Some optimization possibilities are:

--optimization of the firm's management system: the informational, decisional, and management methods and techniques subsystems;

--optimization of management functions; foresight, management, decision, coordination and control;

--optimization in the firm's functioning and development;

--optimization of the company's activity.

Resorting to classical methods of optimization (linear, multi-objective and parametric programming, the transport problem, the multi-attribute decisions, the algorithms in the graph theory, dynamic programming, 'divide et impera' algorithms, 'greedy' algorithms, stochastic programming etc.), non-classical methods (neuronal calculation, evolutionary calculation, fuzzy algorithms, etc.) or very good calculation programs, intelligent systems (neuronal, decision support and expert systems etc.), according to the actual conditions of the Romanian local and global firms, will lead to an increase of the decisional process quality and to the firm's management optimization as well as to the increase of the firm's competitiveness.

There are at least three steps to solve a management problem using a mono-criterion or multi-criterion algorithm /an optimal or rational program:

--The optimal / rational choice of the algorithm class (determinist, stochastic, fuzzy, genetic algorithms etc.);

--The optimal / rational choice of the algorithm from the chosen class of algorithms;

--The optimal / rational choice of the programming language in order to program the chosen algorithm (the enforcement time of the program on the computer, the necessary computer memory, the portability of the realized program, etc., can be considered efficiency functions). According to the probability that an element may or may not belong to a multitude and the degree to which an element belongs to a multitude, the mathematic models are: deterministic, stochastic and fuzzy (vague).

The deterministic models are characterized by variables and secure, achievable relations using the 1 probability and are exact.

When using stochastic (probable) models some variables that describe the studied process are random variables, thus variables have certain values with certain probabilities.

The fuzzy models (vague) use the fuzzy mathematics with variables that cannot be characterized neither exactly (deterministic) nor probably, one can only quality evaluations of their values.

The problem of distinguishing between random (stochastic) and fuzzy (vague) emerges. The random phenomenon results from the uncertainty that an element belongs or not to a multitude, with different probabilities. In case of a fuzzy (vague) phenomenon there are degrees for an element of belonging or non-belonging to a multitude and it is not based on the excluded third from the bivalent mathematic logic. The fuzzy phenomena are based on fuzzy logic, which is a continuous logic that includes the bivalent and n-valent discrete logics.

The decisions in multi-criteria decisional processes with a finite number of variants are called multi-attribute decisions. A multi-attribute decision problem is given, with m decisional variants and n evaluating criteria for variants, m [member of] N, m [greater than or equal to] 2, n [member of] N, n [greater than or equal to] 2. (b. Cociu, 1999)

A. The multi-criteria choice of firm activity field and placement

Choosing the activity field and the firm placement can be made in more than one ways: without considering any optimization criteria, using the intuition and experience, and decisions based on a single criterion, thus mono-criteria decisions, employing the multi-attribute decisions.

A1. Choosing the activity field can be looked at as a solvable multi-attribute decision problem in at least two stages. In the first stage the decisional options can be considered: industrial, agricultural, commercial, transportation, banking and service business etc., and the evaluation criteria may be: market size, competition, estimated profit, competence in a certain field, available capital etc. The optimal option determined in the first stage is detailed in the second stage at definite business fields and those will be the new decisional options in a new multi-attribute problem, and the criteria may be the same ones or others. By a multi-attribute algorithm the optimal business variant is determined (from this stage) from the considered n criteria point of view, other stages to detail the activity field remaining possible. (a. Cociu, 1999)

A2. The firm placement problem is very important, as it involves a great investment, and the misplacement cannot be rectified without important loss. The ideal location is the one that allows the lowest cost for the firm's activity. The firm placement can be made on a multi-criteria base in two stages: in the first stage the region from Romania is chosen: Alba, Arad, Bihor, Timis etc., that represents the decisional options, considering as criteria: placement cost, necessary capital, labor force availability a etc., possibly with certain importance coefficients.

In the next stage a few locations in the chosen geographic area are considered, and the options might be: a city, another town in the same area, different localities etc. The evaluation criteria for the options are: the placement cost, the utilities cost, the existence of roads, railroads, airport, electrical power, water and air quality, competition stage. It can be observed that there is a problem of multi-attribute decision, which is solvable if a specific algorithm is used.

If the firm's location has been decided, another optimization problem refers to the actual placement options: renting, buying a building, buying land and building a facility. The evaluation criteria for the decisional options can be considered: the placement cost, the period of time assigned for placement, the placement duration etc. A problem of multi-attribute decision is obtained, which is solvable with a multiattribute algorithm.

B. The optimal selection of the CAD/CAM/CAE system is a very important problem. A company can make a mono-criteria or multi-criteria selection of the CAD (Computer Aided Design) system, CAM (Computer Aided Manufacturing) system and CAE (Computer Aided Engineering) system.

The mono-criteria selection of the CAD/CAM/CAE system is realized by choosing the optimal variant from among several possible by using a single criterion, for example the price of the system, that is mono-objective decision.

In the multi-criteria selection of the CAD/CAM/CAE system it is considered a multi-attribute decision problem, with m decisional variants and n assessment criteria of the variants.

For example, the following decisional variants are considered:

[V.sub.1] = CATIA system;

[V.sub.2] = SIGRAPH system;

[V.sub.3] = EUCLID 3 system;

[V.sub.4] = Pro/ENGINEER system;

[V.sub.5] = MECHANICA system.

The assessment criteria of the variants can be:

[C.sub.1]--price of the system;

[C.sub.2]--language in which the system can be used;

[C.sub.3]--time of supply;

[C.sub.4]--Conformation exchange with other CAD system;

[C.sub.5]--after-sale service.

It results the consequence matrix

A = ([a.sub.ij]), i = 1, 2, ... ,m; j = 1, 2, ..., n; m = 5, n = 5, and by a multi-attribute decision algorithm: maxmin, Onicescu, Topsis, Electre, Leader, Promethee II, Smrt (Simple Multiattribute Rating Technique) etc. is determined the optimal CAD/CAM/CAE system variant for the company which wants to buy it. Obviously, other decisional variants can be taken into account (CAD/CAM/CAE systems) and other assessment variants. (a. Cociu, 1999)

C. The development of a firm, its expansion in the business, can be made by two complementary ways: internal and external growth. (a. Cociu, 1999)

Choosing the growth solution can be looked at as a multi-criteria optimization problem, in the form of a multiattribute decision problem, with decision variants already mentioned, for example:

* [V.sub.1]--the internal growth of the firm;

* [V.sub.2]--the external growth of the firm that buys shares from another firms (the options are: one or more firms, and if there are more firms , for each firm whose stocks are bought, another decisional variant is considered);

* [V.sub.3]--the firm buys entirely another firm , thus absorption;

* [V.sub.4]--external growth by creating new firms. Other firm development options may be considered.

The criteria for choosing the options may be: the firm growth cost, the estimated profit, the firm's independency, the period of time for the firm's growth etc., and each criterion may have attached an importance coefficient (Andreica et.al., 1998). The optimal solution and the options top are determined through a multi-attribute decision algorithm, for example: Electre, Topsis, Leader, Onicescu, maxmin, Promethee II, Smrt (Simple Multiattribute Rating Technique) etc. (Filip, 2007), (Ionescu et al.,1999).

3. CONCLUSIONS

Using some optimization methods in conceiving, designing, exploiting and developing the Romanian firms, along with management methods, leads to increased performances and values for the firm, in a dynamic competition environment.

A Decision Support System (DSS) may be designed and implemented to solve such decision problems and other structured and unstructured problems that are specific to the Romanian economic environment.

The research results can be applied for the placement and development of a real Romanian firm.

4. REFERENCES

Andreica, M; Stoica, M & Luban, F. (1998)--Metode cantitative in management, Quantitative methods in management, Editura Economica, ISBN 973-590-027-0, Bucuresti

a. Cociu, N. (1999)--Optimizarea deciziilor in sistemele de productie,Decisions optimisation in the production systems, Editura Eurobit, 973-9441-65-3, Timisoara

b. Cociu, N. (1999)--Optimizari in conceperea si exploatarea sistemelor de productie, Optimisations in conceiving and exploitation of production systems, Editura Eurobit, 9739441-65-15, Timisoara

Filip F. Gh.(2007)--Sisteme Suport pentru Decizii, Support systems for Decisions, Editura Tehnica, ISBN 978-973-31-2308-8, Bucuresti

Ionescu, Gh; Cazan, E & Negruta A. (1999). Modelarea si optimizarea deciziilor managerial, Modelling and optimisation of managerial decisions, Editura Dacia, ISBN 973-35-0950-7, Cluj--Napoca.