Master Schedule Creation

NAME

INSTITUTION

DATE

Introduction

The demand forecasting is a process of prediction of future sales by the application of historical data of sales in the making of informed decisions about operations of the organization. This forecast can help in the inventory planning, running flash sales, meeting customer needs and warehousing needs. It is the prediction pf the estimates of the probable demand of product or service in future. It is based on the analysis of the past demand or product in a given market condition. (Suganthi & Samuel, 2012) This should be conducted in a scientific basis and the facts and events that have relation to the forecast should be considered. We focus on the particular product we want track instead of the whole line of products. This makes it easier to organize past data and predict the demand easily.

The need of demand forecasting; it is used in the planning and scheduling of a firms production. It helps in acquiring inputs and providing provision for financing. It also helps in formulating pricing strategy. (Taylor, 2003) It helps in planning the organizations advertisement. Demand forecasting entails predicting the future demand for the firm’s products.

Steps in Demand forecasting

· Specifying the objective.

· Determination of the time frame for the forecast.

· Making choices of method demand forecast.

· The collection of data and data adjustments.

· Estimation and interpretation of results.

Question one

Delphi Method of qualitative forecast

This is a process that is applied to ensure arrival at a group decisions or opinions by the survey of the panel experts. The experts are given several rounds of questionnaires to which the responses are aggregated and this is shared with a group after each and every round. This is an iterative group process that continues until a consensus is reached. This has three participants that includes decision makers (this are people who evaluate responses and makes decisions), staff (the staff administers the questionnaire and the forecast process) and respondents (are people who can make valuable judgements) this is forecasting process framework with a basis on the result of several rounds of questionnaire sent to panels of expert.(Linstone & Turoff, 1975) Multiple rounds of the questionnaires are sent to the panel of experts and anonymous results are aggregated, this is later shared to the groups after every round. This method entails a structured communication technique developed as systematic, interactive prediction style that has reliance on the panel of experts. This technique can be adopted in the use for face to face meeting, this make it to be referred to as mini-Delphi.

Conducting the Delphi forecast

The responses will be summarized and then recirculated to the experts for further comments on the 24 months demand of the products range. This will help in the determination of the range opinions, in testing questions and exploring the demand and achieving consensus on the expert’s perception of the demand. (Skulmoski & Krahn, 2007) The participants will be selected based on their expertise and understanding on the demand of the firm. We will then supply them with a series of questionnaires known as rounds. The ideas in round one will be collated to construct the survey in the next round. The evaluation phase will entail the panelist provided with the responses after which they are asked to reevaluate the original responses. This is interested in the formation or the exploration of the consensus.

Preparation face; here the questions and the topics is established and the questionnaires designed. The participants will be recruited based on their knowledge of demand forecasting. Round 1; the first questionnaire is administered and data is collected and analyzed.

Round 2; we will prepare the second questionnaire based on the first, and then collect the relative data.

Round 3; Prepare the third questionnaire based on the second questionnaires responses, the participants will then rank the statements. We will then consider if more faces are necessary, if so we will proceed with the next face. We will reduce the number of statements and repeat the previous procedures and the final consensus will be collated.

Question 2

a) Mean Absolute Deviation (MAD) we will use mad to measure the accuracy of our prediction by averaging the assumed error or the absolute value of each error. This helps in the measure of prediction error in similar units as the original series. This can be the difference between actual values and the averaging values. This is used in the calculation of demand variabilities. Average distance between each data values and the mean and gives us

The formula; Mean Deviation = (Σ|x − μ|N)/n

Σ is Sigma, (meaning to sum up) || (the vertical bars) mean Absolute Value,

X represents each value in the sense of spread out of the data values.

The data set that have lesser or smaller Mean Absolute Deviation has the data values which are closer to mean as compared to a data set with greater MAD. (Kono & Yamasaki, 1991) Since the data set having smaller MAD have data values close to mean, we will need to update our forecast for jones, since the forecast based on the linear regression has higher average as compared to the one for the first forecast when compared to the actual demand. This is shown below. We will need to adjust the forecast for jones in January.

b) Linear Regression.

Memorandum

To: Amanda Jones (Jones Company)

From: Michel Cory

We will use the linear regression in excel to predict the demand for the remaining months and the following year. We will use the actual demand for the first six months to help us predict the demand for the next 18 months. Linear regression usually tries to model relationships that is between the two variables by the fittings of a linear equation to the data that we have observed. (Seber & Lee, 2012) One variable is the dependent variable and the other independent or explainable variables. In this case we are relating the demand to the 24 months periods. We have the actual demand for the first six months and the predicted demand. Before fitting the linear model to our observed data, we first determine whether there is a relationship between variables that are of interest to us. Based on the graph we find that there are some significant association between our variable. (Montgomery & Vining, 2012) We will plot a scatter graph to show us the strength of the relationships between the two variables. Fitting a regression for the data that is not having an association between dependent variable and independent variables, will not provide a useful model.

Our equation is as follows; y=3.3783x-145499

Our R squared =0.4917, when we user regression analysis a higher r squared is better to explain changes in the outcome variables.

This is in the case the scatter plot doesn’t indicate decreasing or increasing trends. The linear regression line has an equation on in the form Y = a +bx. X will represent independent variable while Y will be the dependent variable. The slope of the line will be (b) while (a) will be the intercept, (The value of Y when x=0)

To get our forecast for the 24 months duration, we will list the months from the first that is January 2019 to the last month in the second year, which is December 2020. We will then feed the actual demand that is for the first six months. Instead of using the linear regression equation that is cumbersome in excel. I will opt for the linear regression forecast formula in excel or the trend formula. In this case we used the forecast formula below.

This will enable us to find the forecast for the remaining 18 months. The excel sheet below shows the forecast of the demand, based on the forecast formula in excel.

Yours Faithfully,

Michel Corry.

References

Konno, H., & Yamazaki, H. (1991). Mean-absolute deviation portfolio optimization model and its applications to Tokyo stock market. Management science, 37(5), 519-531.

Linstone, H. A., & Turoff, M. (Eds.). (1975). the delphi method (pp. 3-12). Reading, MA: Addison-Wesley.

Montgomery, D. C., Peck, E. A., & Vining, G. G. (2012). Introduction to linear regression analysis (Vol. 821). John Wiley & Sons.

Seber, G. A., & Lee, A. J. (2012). Linear regression analysis (Vol. 329). John Wiley & Sons.

Suganthi, L., & Samuel, A. A. (2012). Energy models for demand forecasting—A review. Renewable and sustainable energy reviews, 16(2), 1223-1240.

Skulmoski, G. J., Hartman, F. T., & Krahn, J. (2007). The Delphi method for graduate research. Journal of Information Technology Education: Research, 6(1), 1-21.

Taylor, J. W. (2003). Short-term electricity demand forecasting using double seasonal exponential smoothing. Journal of the Operational Research Society, 54(8), 799-805.