Quality Demand Forecast To Avoid Excessive Inventory In African Fmcg

1) From article: Impact of Poor Forecasting Accuracy:Gross Margin and Organizational Effects of Poor Forecasting Accuracy

Forecasting is an integrated exercise that need the participating of all levels of supply chain to share information in order to improve the forecast accuracy and the performance of the supply chain. Those companies which have good forecasting also improve cross-functional trust, cross organizational trust with suppliers and retailers.

The company should monitor the forecasting accuracy because it’s significant influence to downstream. Firstly, a poor forecasting directly affects the inventory and revenue of the company. For instance, a over-ordered inventory can impact on a company’s cash and working capital. Besides, if the products are out of the market and there is excess material or finished goods in stock which definitely impact on a company’s cash.

At the same time, a stock-out inventory can let the company loss sales as they can not meet the demand of the customers. Secondly, a poor forecasting deteriorate the relationship between the company and its supplier. For instance, the supplier totally reply on the forecasting data from the company, but the supplier won’t follow the company’s data as the company always give a poor forecasting which lead to the overstock in the supplier.

2) From article: Relationship between Inventory Management and Uncertain Demand for Fast Moving Consumer Goods Organisations.Paper presented for 14th global conference on substainable manufacturing

A questionnaire done by 255 personnel comprising of top management, middle management and operational staff within FMCG organisations. The finding shows that there is a positive significant relationship between inventory management and uncertain demand.

Uncertainty causes great challenges in production planning and control. The inventory system consider the demand and supply which is certain or uncertain, and the demand changes at all times caused by the changes on orders. The lower the uncertainty, the better and well controlled in inventory management. The higher uncertainty may lead to stock-out or recessive inventory. To overcome the threat of stock-out by increasing the reorder point by putting extra stock which is safety stock. To lower the inventory, the orgainisations should consider implementation of effective demand and forecasting techniques.

3) From article: Forecast accuracy in demand planning: A fast-moving consumer goods case study Citation metadata

Demand management is to prepare for the right amount of products to customers in the right time at the right price. The FMCG industry usual use push-manufacturing structure as the natural of the product. Therefore, the FMCG industry relies on the forecasted demand figures heavily as all downstream processes refer to this data, such as inventory management, procurement, production planning, resource and transportation planning. At the same time, the flow of information from others (internal, include marketing department, customer service) are important to the development of the demand forecast.

Forecast accuracy has huge influence on inventory investments. Inaccurate forecast may lead to excessive inventory, high logistic cost and loss profit, or stock-out and loss customers. To improve the accuracy of forecasting, the organization should track and analyse forecasting errors and to understand when and why this error occurred. The margin of error and the potential impacts of forecasting error need to known so that some action plans can be made.

The research object company A is a fast moving consumer goods manufacturer in Africa which testify what a demand plan can make a forecast more accuracy and how a forecast accuracy in demand planning can benefit the supply chain.

Firstly, company A set up two objectives for demand planning. The primary objective is to determine the effectiveness of the demand planning intervention of Company A on the supply chain’s performance. The second objective is to determine and compare the pre and post intervention demand planning activities and performance of company A. Secondly, classifying the important product items. Company A classify all products into three types by using ABC analysis which is helpful for forecasting. Thirdly, use techniques and IT tools to assess the forecast accuracy. The techniques like mean absolute percentage error(MAPE) is applied and Torch econometric modelling software and other suitable information technology is implemented to support these intervention activities. Finally, the information interchange and cooperation between demand planner and other department in the company (marketing, customer team and production team).

The findings after intervention activities:

1. The forecast accuracy for all SKUs is improved from pre-intervention period to the post-intervention period.
2. A relationship exist between the improvement in the demand planning process and supply performance. There are positive impacts on the supply chain because of improved demand forecast accuracy. The whole supply chain performance has improved, such as customer service increased, logistic cost getting lower, the ranking of customer service of the company has upgraded.
3. Identifying problems in forecasting consumer demand in the fast moving consumer goods sector.

How to tackle during the forecasting?

The ability to forecast the needs accurately which can ensure product availability without overstocking and overproduction is getting more and more important for both retails and suppliers in fast moving consumer goods industry.

There is 48% of food companies over 50 food companies had indicated that they were poor at forecast and they agreed with importance of forecasting to the company and describing it as a key and critical process. The major gains from effective forecasting:

1. Increased product availability to the consumer
2. Lower inventory levels along the supply chain
3. More effective use of current capital assets
4. Clearer identification of future capital needs
5. True customer/supplier partnerships

How to identify the problems in forecasting consumer demand? First of all, it is essential to understand the forecasting process. The consumer demand is identified as the main input into the process while the output is satisfied consumers, retailers and suppliers. The input and output are affected by a range of mechanisms and controls. Though, there are four issued are identified as greatest concerns to the daily functions.

1. Three forms of Communication—Internal communication, external communication and technology advance.
2. Organization—The organisational structure is not complete and staff lack of training in forecasting.
3. Information is very important in forecasting process. Past history demand of the product can be used to studied for it’s patterns.
4. How to generate the forecast in promotion planning, purchase the forecasting software and measurement of forecast accuracy. Firstly, a better partnership approach to promotional planning between suppliers and retails. Secondly,the orgainisations should purchase the right forecasting software by right person. Thirdly,the measurement of forecast accuracy is essential to monitor the performance of the forecasting method and the forecasting team.

5) From article: Forecasting techniquies in fast moving consumer goods supply chain: A model proposal.

This article presented five forecasting techniques in fast moving consumer goods supply chain. Five forecasting methods are Moving Averages, Exponential smoothing, regression analysis, Estimating trends 1st difference and Estimating trends 2nd differences.

Moving Averages is working well with deseasonalized data while the seasonal patterns goes without trends. It uses the average of a defined number of previous periods as the future forecasted demand.

Exponential smoothing refers to a set of methods of forecasting, there are several popular methods as representative, Brown’s double, Holt’s two parameter and Winters’ three parameter EXPO.

Regression analysis is general approach for modelling the casual relationships between one dependant variable and independent variable.

Estimating trends with difference is applicable if a trend is an increase or decrease in a series that persists for an extended time. Trends are hard to detected when a series has significant randomness and seasonality.

First differences are Yt – Yt-1= change from period t-1 to t(first difference)

Forecasting with differences:

Yt = Yt-1 +b( b means of the differences)

Yt = Yt-1 +b

Yt+m = Yt + m* b(m means number of period)

The process of differences can be used to forecast nonlinear trends using either multiple differences or logarithms. Logarithms are useful when the trend is a percentage growth function and double differences are useful when modelling quadratic functions.

The formula for using second differences is:

Yt-Yt-1 = first differences

(Yt-Yt-1)- (Yt-1-Yt-2) = first differences of first differences= second differences

In the forecasting form, the process of second differences is:

Forecasted amount = 2Yt-1 – Yt-2 + b (b = the mean of the second differenced series and represents a trend estimate)

ReACT system test these 5 techniques and found out exponential smoothing and regression analysis has low error rates.