Data Bases • Custom Term Papers • Free Term Papers • Free Research Papers • Free Essays • Free Book Reports • Plagiarism? Links • Top 100 Term Paper Sites • Top 25 Essay Sites • Top 50 Essay Sites • Search 97,000 Papers @ DirectEssays.com • Search 101,000 Papers @ ExampleEssays.com • Search 90,000 Papers @ MegaEssays.com Free Essays • Term Paper Sites • Chuck III's Free Essays • Free College Essays • TermPaperSites.com • My Term Papers • Get Free Essays • Essay World • Planet Papers	Search Lots of Essays Back to Subjects - Business How OR can Aid the complex problem of Management Decision Making How OR can Aid the complex problem of Management Decision Making How OR can aid Management Decision Making Modern businesses have more need to predict future operations than those of the past do. Managers in large corporations have to summarise and analyse the various data available to them when making decisions. The U.K. OR Society defines the operational research decision-making techniques “as a scientific model of the system, incorporating measurements of factors such as chance and risk, with which to predict and compare the outcomes of alternative decisions, strategies/ controls.” The purpose of these techniques is to help management determine its policy and action scientifically. ‘The models of OR are symbolic or abstract representations of real life problems.’ Examples of techniques that can be used by managers for use in decision making are for example, forecasting. Statistical forecasting is to an extent, an extension of the prediction of a dependent variable. A reasonably accurate forecast can be extremely valuable for a marketing or production strategy. ‘Time series forecasting attempts to capture the past behaviour of the time series and uses this information to predict future values. No external predictors are considered.’ (Kvanli et al, chapter 17) The types of factors that determine the strengths of forecasting are the time horizon of the forecast; the stationarity of the data and the presence of trend, seasonality or cyclical activity. The accuracy of forecasting can be measured by calculating the MAD, MSE and the MAPE. These are useful for comparing the accuracy of a particular forecasting technique on two different time series. The advantage of forecasting is that there is no need to search for external predictors to explain the behaviour of dependent variables. The main disadvantage is that the observed values can be extremely complex and difficult to work out. Such methods are often hard to sell to managers who may not be able to understand the technique. However as mentioned earlier, if a technique is reasonably accurate, it is invaluable to managers. Networks are another decision-making technique that concerns the planning and the control of specific projects. The aim of networks is to complete the project in the shortest time, using the least resources with the minimum cost. Methods that can be used are CPA (critical path analysis) and PERT (Project Evaluation and Review Technique). The advantages of networks for management decision making are that ‘they provide a logical picture of the layout and sequence of a complex project. They also help identify the activities and events that are critical to the entire project. They provide a basis for working out times, costs and resources involved in the project.’ (Drucker, P). They act as a focus for action and co-ordination by the management when performing the project. Finally they make an enormous contribution to both the planning and especially the control of complex projects. The disadvantage is that they do not take into account external factors, for example competition. Another method used in decision making is simulation. This is one of the most widely used. It consists of developing a model of a project, for example in corporate planning. Simulation mimics reality, for example flight simulators and business games. There are many varieties of simulation that are used in operational research, for example, when modelling queuing, one can use discrete event simulation, this is simulation, which models a system as it goes through time. An example of a queuing system is customers waiting at a petrol station, or phone calls arriving at an enquiry centre. The benefit of simulation to management decision making, is that they can ‘model random events based on standard and non-standard distributions and it can predict the interactions between these events, and in turn the knock on effect of an event can be modelled.’ The parameters on the model can be changed and the responses predicted. Therefore simulation can reduce risk, give greater understanding of situations. It can reduce the operating costs, the capital costs and improve customer service.’(Stewart Robinson, 1999) It is preferred to real-life experimentation because experimentation can be costly, however you can change the inputs of a real life system until a satisfactory level has been achieved. For example, fire or hire staff, but this is a good example of why it can be costly, for example if you increase staff, the cost is high. With simulation there is only one extra cost, the time that it takes to change the model. Repeatability is also an important advantage, they can repeat the same sequence of events or alternate scenarios tested under the same conditions. With simulation they can also receive results in minutes whereas real-life experimentation may take a month. The limitation of simulation is that it is only a decision support tool and it doesn’t find exact optimum solutions. A fourth method of OR is linear programming. This includes optimisation, which helps you find the answer that yields the best result. Dependent on your requirements, ‘it is the result that attains the highest profit, output or happiness, or the one that achieves the lowest cost, waste or discomfort. This can aid management decision making because basically it can determine the optimal allocation of scarce resources.’(Lindo website) Linear programming has found practical uses in all different types of field, from advertising to production planning. Examples of use are for problems in the transportation and aggregate production areas. The petroleum industry was an early user of linear programming for solving fuel-blending problems. In optimisation problems, management has to think of the problems as having two important characteristics. The first is limited resources, such as land, plant capacity and sales force size. The second is activities, such as “produce product a” or “produce product c”. Each activity consumes or possibly contributes additional amounts of the resources. The relevant problem is determining the best combination of activity levels that does not use more resources than are actually available. A more specific example of how linear programming using a computer program such as Lindo can be used in practice can be widely seen. This is because using computers, as with other OR techniques increases the ability to apply the techniques to real-life problems. Techniques such as Lindo have a long history of use in industries such as the international beer industry. As quoted on the Lindo website, they have many uses for example, ‘Efe Beverage Group has a 75% share of the Turkish beer market. With an annual per-capita rate of about 10 litres, Turkey’s beer consumption lags many countries of the world. However, Efe's knows the segment of the population that consumes the most beer, 18 to 45 year olds, is forecasted to grow substantially in the next two decades, and they want to be prepared to satisfy this burgeoning group of beer drinkers. To accomplish this, Efes commissioned a study to evaluate sites for new malt plants. LINDO and a mixed integer model were used to determine where to locate the new plants as well as the amounts of barley and malt transported among different locations each year. The objective was to minimise the long-term discounted cost, which included the cost of opening new plants and transportation costs of barley and malt. “Prior to this study, locating a new malt plant at a port town and investing in a private harbour (or at least private loading and unloading equipment) was seen as a viable alternative,” remarked Serdar Bölükbasi, Marketing Director, Efes Beverage Group. This would allow the company to avoid certain costs associated with loading and unloading the barley and malt. (“The economic analysis conducted in the study clearly demonstrated that this alternative was far too expensive relative to its benefits.”). In conclusion, the main advantage of OR models is that they provide a basis for the solution of complex problems in static/ dynamic situations. Such models can be designed to take a large number of factors into account at any one time, for example linear programming. They can reduce these factors to mathematical terms and experiment with them by introducing a variety of inputs to assess what effects they have. All this can be done without interfering with the operational/planning processes currently under way in the organisation. Then the risks of a particular strategy can be evaluated in a relatively safe manner, before being put to the test in a real life context. The disadvantages of OR are that the techniques can take time to be developed, therefore they are less useful for producing quick answers. These methods can also represent an oversimplified picture of a particular set of conditions and therefore may suggest only partial solutions. Line managers may also resist them on the grounds that such models are too theoretical to be put into practice. However, what has come to my attention, is that these limitations, have to some extent been reduced by the wider use of computers to handle complex calculations, and the development of OR teams made of managers from a variety of disciplines. The accuracy of these decision techniques is also dependent on the area that they are being used in. For example, the accuracy and reliability in production is more accurate than in marketing. This is due to control, most of the relevant factors in production are under control by decision-makers, whereas many crucial factors are outside control in marketing. Bibliography: References Drucker, P; Drucker on management, London : Management Publications Limited for the British Institute of Management, 1970. Lucey, T,Quantitative Techniques Word Count: 1540
Copyright © 2005 College Term Papers, INC All Rights Reserved.