The most important part of the "Operations Research" enterprise-forecasting and decision-making

Operations Research is a science that studies how to organize and manage human-machine systems effectively. In the field of management, the measurement method of decision-making in management decision-making. In the plan, winning thousands, the main responsibility of the enterprise leader is to use the theory knowledge of operational research, combine the various environment of the enterprise, then make the decision, select the best plan, help the enterprise realize the benefit maximization.

Everything is pre-set, not pre-waste. Forecasts are estimates and judgments of uncertain future events. Forecasting is the basis of decision making, which provides data and information for enterprises to make decision. Enterprises should constantly according to the changes in the external environment, the development trend, the impact of national policy changes to predict, prediction is a study of the future and explore the future of the subject, if we know through the prediction of changes in the law of business, then will create revenue for the enterprise.

The prediction is mainly based on different situations and using reasonable methods.

Judgment Prediction Method:

Delphi method: We hope to get a unanimous opinion in the "Expert Group", which features: 1. Comments are anonymous. 2. The information is repeatedly fed back and discussed. 3. Finally, the research department to collate and summarize the views of experts.

Expert Group law: As the name implies are experts, face-to-head consultation. But it is easy to be swayed by the ideas of the authorities to discuss the results. It is suitable for short-term forecasts.

Time Series Prediction Method:

Understanding is easier, there are arithmetic mean predictions and weighted predictions, and each prediction is divided into horizontal and vertical.

Exponential smoothing prediction is a combination of qualitative and quantitative methods. The predicted value of the next period is calculated by formula, smoothing coefficient, predicting value and actual value.

Regression model:

A simple linear relationship between an independent variable and a dependent variable. The key is to determine the parameters of the regression model, using the least squares method.

After the prediction, a situation may be a variety of feasible solutions, and the specific implementation of which, how to achieve optimal, need the enterprise leadership comprehensive consideration, after discussion, make decisions. Decisions are made for execution, purposeful and targeted. A broad range of decisions consists of four programs: 1. Clear purpose. 2. Develop a number of viable options. 3. Use predictive methods to predict and derive decision matrices. 4. Optimal solution.

Decision-making is purposeful, decision-making is not static, different environment, decision making is different, there are three kinds: under the condition of determination, uncertainty and risk condition.

Decision making under uncertain conditions

Maximum maximum:"big in large" optimists decision-making criteria, choose the maximum value from each scheme, choose the largest from the maximum value of all schemes.

Maximum minimum:"small Medium big" conservative decision-making criteria, select the minimum profit value for each scheme, choose the maximum from the minimum return value of all schemes.

Min Max: based on regret value, "Big medium take small"

1. Select the same scheme, the maximum value in different states.

2. In the same state, use the maximum value minus each scheme. A table of regret values is drawn.

3. List the maximum regret values for each scenario.

4. Select a minimum value for regret.

Realism: The best probability of a situation is a, the worst bit (1-a), sum by probability, and then select the maximum value of the gain.

Risk conditions

In this case, it is necessary to determine the probability values of various states and enhance the scientificity. The decision under the risk is based on the expectation standard.

Maximum expected income: (form of decision-making income statement)

1. Determine the probability

2. Conditional profit: Draw out the form and make it clearer.

3. Expected profit: Sum the situation according to probability and add profit.

4. Maximum expectation of accurate intelligence.

Minimum expected loss:

Can be converted by the maximum expected value. The maximum benefit of each state minus the benefits of other schemes. The corresponding table is subtracted from the column values, the table has a loss of 0 diagonal, the upper is the opportunity loss, the lower part is the scrap loss.

Enterprise's decision-making, is to build on the basis of the forecast, the forecast has the qualitative and quantitative points, they each have characteristics, the specific implementation under different circumstances, different enterprises use the way of forecasting is different, need to take into account the structure of the enterprise, scale, the length of time to create. Decision-making is the analysis and execution of the result of the forecast, and realizes the optimal benefit.

The most important part of the "Operations Research" enterprise-forecasting and decision-making