In the following, we demonstrate how to get the most out of Datup forecasts to increase accuracy levels and decrease lead times in demand planning. To do this, the basic, intermediate and advanced journey and usage scenarios will be described.
These use cases will allow the analyst to take into consideration various fields of the results cubes, which will better guide decision making with respect to sales, demand, supply or similar behaviour.
Basic Use Case: Suggested Forecasts
This use case guides better decision making for:
Know the volume of demand required by customers for each item for one or more future months.
Propose in S&OP sessions demand volumes according to optimistic, conservative or pessimistic sales conditions.
Assess the certainty or uncertainty of demand forecasts.
Prioritise portfolio forecasting by top-selling items, turnover and demand variation.
This scenario focuses on suggested forecasts, automatically generated and selected by the platform. First, you must locate the date and the item of interest, by means of the columns
Itemrespectively. The column
Pronóstico Sugerido shows the suggested forecast for the date and item under analysis, based on historical demand behaviour (
Demada Histórica); considering seasonality, levels, trends and even relationships with other items in the portfolio.
In addition, the column
Pronóstico Sugerido Supwhich presents an alternative scenario to the suggested forecast when there are indications or a substantial increase in demand for the item is anticipated for the specific period. The
Pronóstico Sugerido Sup is often very useful for S&OP sessions, where the trading counterpart often arranges higher values for forecasts of certain items, subject to marketing campaigns or higher consumption strategies. Instead of opening the space for speculation for such a possible higher level of demand, the platform offers an alternative value that contemplates such an over-demand scenario. On the other hand, the solution also presents an alternative value for the under-demand scenario, where the trading partner foresees a discouragement in the consumption of the item for future periods. The field
Pronóstico Sugerido Inf contains the amount of demand to be considered.
WMAPEThe error expected for the forecast period, based on the errors obtained by testing the backtestsThis is to simulate the demand forecast for previous periods, where the actual behaviour is known, hence it is possible to measure performance. By default, Datup performs 5 of these simulations, or backtests in each iteration of the models.
Finally, it is suggested to take into account the column
RankingThis allows us to determine the items with the highest consumption, turnover and stability in demand. In other words, it identifies the items of highest and lowest value for the business.
Intermediate Use Case: Suggested Intervals
This use guides better decision making for:
All the advantages of the basic journey.
Know which items are recurrently over-predicted with respect to their historical demand to prevent excessive inventory losses.
Know which items are recurrently under-predicted with respect to their historical demand to prevent out-of-stock losses or stock-outs.
The second scenario accompanies the suggested forecasts with suggested ranges, allowing the demand analyst to know which items have been consistently under- or over-forecast in the past. The column
Intervalo Sugerido indicates the forecast interval associated with the suggested forecast in column
Pronóstico Sugerido. Accordingly, the suggested upside forecasts (
Pronóstico Sugerido Sup) and downward (
Pronóstico Sugerido Inf) are also accompanied by their suggested intervals
Interval Sugerido Supy
Intervalo Sugerido Infrespectively.
The forecast intervals
Intervalo Sugerido Sup y
Intervalo Sugerido Infare not only calculated for the periods to be forecast, but also for each forecast period. backtest or simulation. It is precisely in this way that it is possible to determine what is the forecast interval, and hence the most likely suggested forecast, based on its repetition in the observed demand history.
As a practical matter, the demand analyst or planner can determine whether an item is susceptible to over-demand if the suggested range
Intervalo Sugerido has one of the following values:
Lo60. On the contrary, the item is susceptible to overdemand if the
Intervalo Sugerido presents either
Up95. In both cases, the over-demand is set with respect to the forecast point that lies in the middle of all forecast intervals.
Using forecast intervals allows the demand planning process to anticipate and avoid overproduction or oversupply events, as well as stock-outs or stock-outs.
Advanced Use Case: Naive Forecasting and Forecast Intervals
This use guides better decision making for:
All the advantages of the intermediate journey.
Examine naïve forecasts, i.e. averages or last observations, to plan demand on items with high variation or intermittency and increase their accuracy.
Examine the suggested forecasts for each item and period to see how high or low demand may behave in different market circumstances.
Evaluate the MASE error to determine in which cases it is advisable to use naïve forecasts to increase planning accuracy.
The third and final scenario involves the naïve forecast and all the associated forecast intervals estimated by Datup. This collection of values allows the analyst to know all possible scenarios of high and low forecasted demand.
The forecasts point
Pronóstico Puntolower 95
Pronóstico Inf 95, lower 80
Pronóstico Inf 80, lower 60
Pronóstico Inf 60, higher 60
Pronóstico Sup 60, higher than 80
Pronóstico Sup 80 and above 95
Pronóstico Sup 95 The total number of points Datup estimates for each date and item to consider the different scenarios of high, medium and low demand. The algorithm automatically selects the scenario, and in turn the most likely value. However, the cube results are presented in an open way to the analyst to enable possible additional exercises.
For its part, the naïve prognosis in the column
Pronóstico Naive calculates the forecast value by means of a rolling window average. The size of the window coincides with the number of periods to forecast. For example if you want to forecast 4 weeks, the rolling window calculates the naive forecast by taking the average of 4 weeks at a time from history. Datup compares the performance of its model against the naïve forecast, as there are items whose stability in the demand history suffers from high intermittency and/or random sharing that prevent their forecasting through traditional or advanced statistical methods. The value of the
MASE shows the result of this comparison. Values close to 1 (above or below) confirm the superior performance of the Datup model, while values well above 1 favour the use of the naive forecast.
Pronóstico Naive. If this is the scenario, in the bucket, the naïve forecast will be the value suggested as the forecast for the period evaluated.
Advanced Use Case: Impact, Correlation and Causality
This journey guides decision making to:
Identify the variables or indicators that have the greatest impact on the forecast for one or more items of interest. Upward or downward changes in the associated variables are most likely to produce significant variations in the planning forecasts.
Determine the variables or indicators correlated with the items of interest. That is, the variables whose upward or downward changes produce changes in the Items of Interest, in the same proportion.
Determine the causal variables or indicators for the items of interest. That is, the variables whose upward or downward changes are a direct cause for the future behaviour of the items of interest.
In conjunction with demand planning, it is possible to know the items, variables or indicators that have the greatest impact on the forecasts, accompanied by a percentage of importance; the higher the percentage, the greater the importance of the variable. Datup automatically identifies and orders the determining variables in the planning. In this way, analysts can focus their efforts on tracking key indicators whose changes have a direct and considerable influence on forecasts. In the example in the graph, US Existing Home Sales is the most important indicator for the generated forecasts with a 15% share, followed by the Acetic Acid Index (7.4%), Acetic Acid RMB (6.4%) and US Ethanol Price (6.2%). Thus, analysts can identify among hundreds of variables, which are the top-n that really influence purchase prices, sales quantities or demand units for future periods, in order to put in place better targeted trading, purchasing, production or distribution strategies in advance.
Correlation provides a second criterion in the selection of the variables or indicators that deserve greater attention in the decision-making processes for purchasing, negotiation, production or distribution, as it determines which are the variables whose upward or downward trends generate a direct or inverse behaviour in an item of interest. In general, variables with a direct correlation between 0.7 and 1 or an inverse correlation between -1 and -0.7, deserve to be taken into account. In this case, the item or indicator of interest is the purchase price of the raw material Vynil Acetate Monomer (VAM). The directly correlated indicators are: Price VAM ADC (0.9), VAM Index (0.86), Acetic Acid Index (0.85) and Acetic Acid RMB (0.85) . At this point, special attention should be paid to those variables that appear with high percentages of importance and high correlations, i.e. Acetic Acid Index and Acetic Acid RMB, as their upward variations will most likely produce upward variations also in the price of VAM. It is important to check if there are variables with high negative correlations (-1 to -0.7), as they may indicate an inverse behaviour, a rise in the variable will indicate a downward trend in the Price VAM. In this example, there are no inversely correlated variables.
Finally, causality is the last criterion in the analysis and selection of important variables in price, sales or demand forecasts. Causality, unlike correlation, allows establishing a cause-effect relationship between a group of variables and indicators and the items of interest. Based on the forecasts, Datup determines which are the variables whose behaviours are the cause of upward or downward changes in the target items. For example, it is observed that VAM price forecasts are mostly caused by the Methanol China Index (0.5), New Orleans Temperature (0.39), Acetic Acid Index (0.35), Qingdao Temperature (0.31), US Price Ethanol (0.31) . Once again, it is important to pay attention to those variables common to impact, correlation and causality analysis, as they form the set of key indicators to be taken into account by analysts for their operational and strategic decisions. Here, Price Ethanol is identified in both impact and causality analysis. Even more relevant, the Acetic Acid Index is the common denominator in all 3 significance analyses.
Thus, the key indicators in order of importance to guide Price Vynil Acetate Monomer (VAM) purchasing strategies, following the criteria of impact, causality and correlation are:
Acetic Acid Index
Acetic Acid RMB
US Ethanol Price
US Existing Home Sales
Methanol China Index
New Orleans Temperature
Finally, it should be noted that each new daily, weekly or monthly forecast involves the estimation of high-impact variables and indicators, causalities and correlations, based on changes observed in the most recently collected data. Datup automatically and continuously identifies and presents to analysts the top-n of key indicators to be taken into account for the best decision making in trading, purchasing, production or distribution.
Advanced Use Case: Scenario Simulation
This use case guides decision making for:
Simulate the behaviour of items of interest, based on the increase, maintenance or decrease in the values of specific variables and indicators.
Taking the identified high impact, causal and/or correlated indicators identified by Datup's importance analysis as a starting point, it is possible to forecast the quantity or price of an item of interest in future periods. Unlike the forecasts of the first part, the simulation not only considers historical data, but also builds various scenarios from the key indicators to generate the forecasts of the items of interest; covering increases of 10% and 30%, drops of -10% and -30% or the permanence of the current value. In addition to the indicators postulated by the impotance analysis, analysts can include additional variables to simulate.
In the example use case, Price Vynil Acetate Monomer (VAM) is the item of interest, the key indicators to simulate are: Acetic Acid RMB, US Ethanol Price, US Existing Home Sales. In addition, the supporting indicators Dollar TRM and WTI Crude USD are included in the simulation.
The results show the simulation of 5 scenarios that can be presented by the Acetic Acid RMB in the short-term forecasts, which correspond to increases of 10% ($9,995) and 30% ($11,765), falls of -10% ($8,145) and -30% ($6,335) and the current price ($9,050) remaining unchanged. For each scenario, Datup forecasts the Price VAM most likely in the column
Pronóstico VAMaccompanied by the ceiling price
Pronóstico VAM Sup and floor
Pronóstico VAM Inf where it can be located, depending on the uncertainties observed in its history. For example, if the Acetic Acid RMB increases its current price by 10% to reach $9,995, the Price VAM will stand at $2,309, with a ceiling at $2,337 and a floor at $2,282.
Now, taking the TRM dollar as one of the variables to be simulated, we consider the scenario where the exchange rate falls by 10% to $3,460. In this way, the Price VAM will be $2,246, with a ceiling at $2,284 and a floor at $2,213.