Navigating the challenges: declining portfolio forecastability 

In the vibrant landscape of forecast and demand planning, portfolio forecastability faces turbulence due to various factors. Market shifts driven by geopolitical events, technological advancements, and unforeseen global crises introduce an element of unpredictability that challenges traditional forecasting methods. Evolving consumer preferences, shaped by trends, cultural shifts, and societal changes, add layers of complexity that demand planners must skillfully navigate. External factors, such as economic fluctuations and supply chain disruptions, amplify the challenge, making accurate demand anticipation increasingly elusive. 

The XYZ analysis, a renowned method of categorizing products based on demand volatility, vividly illustrates increased variability and reduced predictability of sales across nearly all industries in the last 12 months. For a growing part of company portfolios statistical approach is no longer sufficient.  More and more items need meticulous manual planning or the adoption of more agile and adaptive inventory strategies.

The path forward: agile and adaptive forecasting approaches 

In turbulent times, the limitations of relying solely on statistical forecasts become strikingly apparent. While statistical models excel in capturing historical patterns, they may struggle to swiftly adapt to sudden market changes. The assumptions underpinning these models—based on historical data and relatively stable conditions—can falter in the face of rapid changes. 

So, how can demand planners respond effectively to the challenge of declining forecastability in turbulent times? It demands a strategic shift towards agile and adaptive forecasting approaches, incorporating not only historical data but also real-time insights, forward-looking statistical methods, and the ability to swiftly adjust to changing circumstances. An increasing number of companies are dipping their toes into driver-based forecasting, a strategic approach identifying and leveraging key drivers influencing demand.  

This method employs machine learning techniques and artificial intelligence to enhance predictive accuracy. Driver-based approach builds on the statistical model foundation, recognizing that not all products or services are influenced by the same factors. By understanding the specific drivers affecting each portfolio component, demand planners can tailor their forecasting strategies accordingly. 

Driver-based forecasting relies on a detailed analysis of the various factors impacting demand for each product or service. These influencing factors can be categorized into both internal and external aspects. In specific industries, external factors such as weather conditions, economic indicators, and market trends play a significant role. Conversely, in different sectors, internal indicators like contract positions and order books carry more relevance. Incorporating these drivers into the forecasting model enhances the accuracy and relevance of predictions.

Difficulties in adopting driver-based forecasting 

The adoption of driver-based forecasting comes with inherent challenges. Organizations are required to invest substantially, leveraging advanced analytics tools that align with this methodology. Equipping demand planners with the requisite skills and establishing a resilient feedback loop for continuous enhancement are integral components of this implementation. Many supply chain managers grapple with the evaluation of the necessity for such an investment, recognizing the complexities involved in this decision-making process. It’s a decision that involves not only financial considerations but also a comprehensive understanding of the specifics and volatility of the demand dynamics, what statistical forecast still brings and where additional drivers would be required. 

In order to help companies to solve this puzzle, EyeOn came with the Fast Forecast Scan. It is a quick tool to provide rapid insights into demand characteristics and forecastability. The Fast Forecast Scan expedites the identification of improvement opportunities, guiding strategic decision-making on proper forecasting methods, be it statistical or driver-based. The scan unveils the highest possible statistical forecast accuracy for each product in your portfolio, identifying key improvement opportunities in your current forecast setup. Our experienced specialists identify items challenging to forecast with statistical methods and that can have potential for further driver-based analysis. Witness the transformative power of the Fast Scan through our on-demand demo.

The post Declining forecastability: Can your portfolio still be forecasted? appeared first on EyeOn.

However, traditional forecasting methods like time series forecasting often fall short of effectively utilizing this data. Planners often have to manually enrich the statistical forecast, which is time-consuming and introduces the risk of human bias.  

Fortunately, machine learning offers a solution by leveraging past and future data to identify demand drivers and enhance demand predictions. But a challenge arises due to the inherent black-box nature of powerful machine learning models. This is where Explainable AI (XAI) methods come into play, offering transparency and transforming the black box into a glass box.

Explainable AI (XAI)

During my research at EyeOn, I explored various Explainable AI (XAI) techniques and their alignment with the stakeholder’s needs in the demand planning process. I discovered that explanations are not one-size-fits-all; they depend on the interaction between the user and the explanation tool. Different stakeholders have unique requirements, which reveals a significant gap between the needs for explanations and the existing algorithm-focused approaches. 

By examining stakeholder needs, I found that model developers seek explanations to improve and refine models, planners require a solid understanding to support their decision-making and persuade others, and general managers seek confirmation of effective model utilization. Recognizing these diverse needs is crucial for successful implementation and acceptance during the planning process. 

Overall, I enjoyed putting these techniques into practice and gaining practical experience. It was nice to uncover each technique’s strengths and weaknesses and better understand different stakeholder’s needs and motivations. With the support of a welcoming and innovative team, I am optimistic about future opportunities and the positive impact we can achieve by leveraging this knowledge and these techniques.

Curious to Learn More? Get in Touch with Our Experts!

If you’re interested in learning more about these findings ,their practical implications or engage in further discussions, feel free to contact one of our experts. Want to know how we use this? Take a look at our driver-based forecasting framework. Together, we can continue advancing the field of explainable demand planning.  

The post Delivering the right message with Explainable AI in machine learning appeared first on EyeOn.

4-minute read 

A snowy predicament: how machine learning models can learn the wrong things

Imagine a machine learning model that’s supposed to tell the difference between a wolf and a husky, but instead, it’s learned to recognize snow. This might sound like a joke, but it’s a real-world example of the limitations of current machine learning models. Researchers have found that a model trained with inadequate data can make mistakes in classifying images of wolves and huskies, and instead learn to identify snow as the main characteristic! The key message here is that for demand planners to trust a machine learning model, we need to have a way of asking the model how it arrived at its predictions.

From wolfs and huskies to machine learning in demand planning

In the field of supply chain demand planning, we’re in the middle of a shift towards the use of machine learning models. At EyeOn we often refer to this as driver-based forecasting. These models offer great potential to provide more accurate and less biased forecasts by incorporating external drivers. However, with an increased use of machine learning also come challenges. One of the main challenges is that these models are often considered to be black boxes, making it difficult for demand planners to understand how the model arrived at its predictions.

To truly reap the benefits of driver-based forecasting, demand planners need to effectively work together with these machine learning models. And to do this, trust is key. But how do we establish trust in a model that we can’t fully understand?

Explainability addresses this challenge by providing demand planners with a clearer understanding of the predictions and decision-making processes of machine learning models. This transparency helps build trust in the models as practitioners can see how the model arrived at its predictions. By having insight into the model’s prediction processes, planners can more effectively collaborate with the model; trusting it when it is correct and recognizing when it might be off. Additionally, explainability can also lead to a deeper understanding of the underlying business dynamics and challenge previously held assumptions.

From theory to practice: explaining the demand forecast

Let’s bring this to life with a concrete example: Imagine you are a demand planner working with machine learning models to generate a demand forecast planning for your portfolio. You have a good understanding of the drivers that influence demand for your products, including the impact of promotions. However, it can be challenging to get a clear picture of how the different drivers interact and affect your forecasts. To validate your understanding, gain trust in the model, and make better-informed decisions, two visualizations can be of great help.

First of all, let’s say you have a demand forecast of 73.64 for a specific product in a certain time period. The ‘waterfall plot’ can show us how the model arrived at this forecast. The plot starts with the baseline forecast, then each row shows the positive (red) or negative (blue) impact of each driver (or feature) on the forecast. In this example, we can see that the forecast is higher than the baseline due to the presence of a promotion for this product and a high number of open orders for the same period.

Secondly, for a higher level summary of the impact of each driver, the ‘beeswarm plot’ is a useful tool. It displays the impact of the top drivers on the model’s forecast in a compact and clear format. Each forecast for a product-time period combination is represented by a dot on each feature row. The dot’s position is determined by the driver’s impact and the dots pile up along each feature row to show density. Colour is used to show the original value of the driver. It becomes evident that the presence of a promotion (represented in red) has a positive effect on the forecast, whereas its absence (represented in blue) has a negative effect. Likewise, the number of open orders is also seen to have a direct impact on the forecast, with a higher number generally resulting in a more positive influence.

Conclusion on using machine learning in demand planning

Machine learning models are rapidly becoming a crucial tool in supply chain demand forecasting. However, their black box nature often makes it challenging for demand planners to understand and trust the predictions made by these models. That’s why explainability is becoming increasingly important, as it provides planners with transparency into the models’ decision-making processes and helps build trust in the results. In conclusion, explainability can act as a catalyst for change in the supply chain planning and forecasting process, as it allows practitioners to challenge business assumptions, learn about underlying business dynamics, and eventually make better-informed decisions.

Are you interested to learn more about driver-based forecasting?

Please contact us to see how driver-based forecasting can be of added value to your organization! Reach out to Rijk van der Meulen or Willem Gerbecks.

The post The importance of interpretability for effective demand planning in the age of machine learning appeared first on EyeOn.

Why loss functions are essential for achieving accurate and reliable predictions from machine learning models in the supply chain context

A relevant consideration in performing time series forecasting using machine learning models is the effect of different so-called ‘loss functions’. Loss functions are the driving force behind any machine learning model. They play a crucial role in evaluating the model’s performance. Loss functions are how one measures the difference between the predicted and true values, and they guide the model during the training process to find the optimal set of parameters – minimizing the loss.

Our intern Dirk Cremers took a deep dive into this topic for his master thesis. Here’s what you need to know:

The different types of loss functions and how they work

You can use many different loss functions for time series forecasting, including the most known ones: the mean absolute percentage error (MAPE), mean absolute error (MAE), and root mean squared error (RMSE). However, in addition to these well-known loss functions, there are other loss functions (e.g. Huber and Tweedie) which can provide improved performance in certain scenarios. In the remainder of this blog post, we will go a bit more in-depth about these loss functions and explain them in more detail. However, before describing each loss function, here are some definitions that we use in this article:

• N is the number of observations
• F is the forecasted value
• A is the actual value

MAPE – the mean absolute percentage error

First off, the mean absolute percentage error (MAPE): This function is commonly used to evaluate the performance of machine learning models for forecasting. However, it is also used as a loss function. The MAPE is formulated as follows:

The main advantage of using this loss function is that it results in a forecast that respects zero values in a dataset quite well (e.g. predicts a zero often when the actual is also zero). This behavior is due to the following: the MAPE divides each error separately by the demand, and is thus skewed: high errors during periods of low demand would greatly affect the MAPE. Therefore, optimizing the MAPE will result in a forecast that most often underpredicts the actual demand. For many businesses, this is not the kind of behavior they are looking for.

MAE – the mean absolute error

The mean absolute error (MAE) is a widely used loss function in machine learning, particularly in forecasting. It is defined as the average absolute difference between the predicted values and the true values, and can be expressed mathematically as:

One of the main advantages of using the MAE as a loss function is that it is relatively robust to outliers. Furthermore, an interesting property of the MAE is that it can be shown to optimize according to the median of the errors.

However, the MAE has also some limitations. For example, it does not penalize large errors as heavily as other loss functions such as the root mean squared error (RMSE). This can be problematic in cases where large errors are more costly or most detrimental. Additionally, the MAE does not take into account the error relative to the actual value. Therefore, this loss function considers predicting 510 when the actual value is 500 as an equal error to predicting 20 when the actual is 10. Most planners would not agree with this judgment.

RSME – the root mean squared error

The root mean squared error (RMSE) is a useful loss function when large errors are more costly and it is important to minimize them. This is obtained by taking the squared root of the average squared difference between the predicted value and the actual value, e.g.

As mentioned, the advantage of this loss function is that it heavily penalizes large errors, which can be useful in many cases. Furthermore, it can be derived that the RMSE optimizes according to the mean of the squared errors, in contrast to the median (MAE).

However, the RMSE has some limitations as well. For example, it is sensitive to outliers in the data, as large errors significantly increase the overall error. Furthermore, just like the MAE, it does not take into account the error relative to the actual value.

The Huber loss function

To combine the best of two worlds, the Huber function was proposed (Chen, 2018). The Huber loss function is namely a combination between the MAE and the RMSE and can be expressed mathematically as:

One of the main advantages of using the Huber loss function is that it combines the properties of both the MAE and the RMSE. Like the MAE, it is relatively robust to outliers, as it is not heavily influenced by extreme values in the data.  Like the RMSE, it heavily penalizes large errors, which can be useful in cases where large errors are more costly. As a consequence, the Huber metric optimizes according to a combination of the median (MAE) and the mean (RMSE) according to this δ value and the size of the errors.

However, the Huber loss function has some limitations as well. For example, it requires the user to choose a value for the δ parameter, which can be challenging and may require some trial and error to find the optimal value. Furthermore, the Huber loss function still does not capture the relative error to the actual value.

To visualize the effect of these four loss functions we’ve created an example with four different forecasts according to loss functions. Furthermore, the example includes some peaks on certain days due to promotion. As one can see, each forecast clearly shows the properties of the specific loss function.

The Tweedie loss function

Finally, let’s discuss the Tweedie loss function (He Zhou, 2019). This loss function was initially introduced for predictions of insurance claims due to its ability to capture whether a zero value is expected. However, because of its ability to capture data that is skewed or with a heavy tail, it was also introduced for forecasting. For example, in the case of the dataset containing many zero values, traditional loss functions such as the mean absolute error (MAE) or root mean squared error (RMSE) may not be effective, as they can produce unreliable results when applied to this dataset.

Recent competitions like the M5 show that in these settings the Tweedie loss function outperforms others and results in a great performance. This is obtained by the Tweedie loss function being a generalization of the Poisson, Gamma and inverse Gaussian loss. Based on a specific value of the Tweedie parameter (p), this loss function can be expressed and aligned with the actual demand as well as possible.

The main advantage of the Tweedie loss function is its ability to capture a wide range of distributions and be tailored to specific types of data by selecting an appropriate value for p. This in particular helps in forecasting products where the prior data exhibit intermittency or bulk purchasing.

However, to use the Tweedie loss function also means engaging in the trial-and-error process of finding the correct p-value for a given data set, which can be challenging.

Conclusion

In general, the choice of loss function for time series forecasting will depend on the specific characteristics of your data and your business goals. It is important to carefully evaluate the performance of the model using different loss functions to find the one that provides the best results. By exploring new loss functions, we can find new ways to measure the difference between the predicted values and the true values, and this can lead to better performance of the model.

Therefore, loss functions are a crucial part of time series forecasting using machine learning models. By carefully choosing the right loss function for the task at hand, you can improve the accuracy of your predictions and make better business decisions as a result.

Do you have open questions or need support in choosing the right loss function for your business? Contact our expert Dan Roozemond or get in touch with us.

The post The heart of machine learning: Understanding the importance of loss functions appeared first on EyeOn.

Why driver-based forecasting is the key ingredient to success

The biggest challenge in forecasting is how to come to a fact-based and realistic plan and forecast. Although in every business there are internal and/or external factors that have a causal relation to demand, it can be hard to bring the right drivers for demand into your forecast. What if your forecasting method would do this for you, so you can focus on more important business decisions? By using demand drivers as the basis for forecasting, you can process and deliver the right information to the right person’s desk at the right time.

How does driver-based forecasting work: Complement your forecast with the right spices

The already mentioned internal and/or external factors can help you foresee how the demand of your products is going to develop. Examples for internal drivers are promotions, orders, or a product plan. External drivers are for example point of sale information, holidays, or the weather.

Traditional forecasting methods, such as time series forecasting, don’t take this data into account. The statistical forecast needs to be manually enriched by your planners. This requires a lot of time and skill from the planners and introduces a risk of human bias.

With machine learning, we can identify demand drivers for your business and learn from the past data of these drivers to enhance the predictive power of your forecast. This approach is much more fact-based, offers an automated way to get insights, and enables the planners to balance time and effort.

The recipe for starting driver-based forecasting

Are you ready to steer your business based on more precise forecasts but don’t know where to start? Creating a value-adding machine learning model can be challenging. In our next blog, we will talk about the hurdles companies might face when implementing driver-based forecasting and how to overcome them.

Can’t wait to start improving your forecast performance? Taste the future with driver-based forecasting: Contact us now!

Or learn more about our Fast Forecast Scan. The EyeOn Fast Scan provides organizations with a clear stick in the ground of current forecast maturity and the potential that can be reached. This can be used as a great starting point for taking the right improvement steps.

The post Taste the future: What is driver-based forecasting? appeared first on EyeOn.

Work smarter, not harder – with reinforcement learning in inventory management

The traditional way to solve this problem would be by using approximations, heuristics, or simulation. We say this is a challenge for machine learning!

Reinforcement learning is a branch of machine learning that trains the computer to make the right decisions. In the beginning, the accuracy is low but as you keep training the results get better. Imagine you are training a dog to bark when a stranger is approaching by giving it a treat when it does it correctly and telling it off when it barks at a friend. With time it learns to identify when to do what.

Reinforcement learning solves your inventory management challenge by finding the most cost-effective policy. This way, you don’t need to evaluate all the policies, saving you time and effort. With reinforcement learning, you can solve complex inventory problems, and yet maintain an interpretable solution and simple decision rules. It can handle not only constant but also changing demand patterns, a major advantage in today’s volatile environment!

We get your business years ahead

To enrich the offering to our customers, we at EyeOn explore and apply the latest methodologies and techniques. The project of our intern Floor van Helsdingen is the perfect example: Floor successfully applied reinforcement learning to a complex inventory problem of one central depot supplying multiple local customer-serving depots. There is hardly a hotter topic these days than machine learning, and we have demonstrated its potential for inventory challenges.

Next to classical reinforcement learning methods, we are exploring deep reinforcement learning. With this innovative method, based on artificial neural networks, we will be able to achieve even better solutions and handle even larger product portfolios.

Curious if reinforcement learning can help optimize inventories in your business as well? Reach out to our experts Maarten Driessen or Dina Smirnov!

The post How reinforcement learning solves inventory management challenges appeared first on EyeOn.

Data is everywhere in today’s world. Every single event in your supply chain generates data, and the opportunities to utilize all that data seem endless. At the same time, there are many tools and technologies out there claiming to solve all your data problems. No wonder many professionals feel overwhelmed!

Which key opportunity can you unlock by using your data better? Reducing latency! Quicker data collection, better dashboards, and suitable tools help you get the right information on the right person’s desk at the right time – effectively reducing the delay from event to decision, and thereby dramatically increasing the impact of your decisions. Machine learning can help.

Results from a benchmark conducted by EyeOn in 2021 show that:

1. companies are most eager to start applying machine learning in the area of demand forecasting
2. only 10% are actually doing this at that moment.

While the goal for each company is straightforward – trying to improve forecast performance with less effort spent by demand planning, sales, and marketing altogether – there is no one-size-fits-all solution that can do the job.

Do you want to improve your forecast performance and accelerate decision-making in the supply chain? Watch our video on driver-based forecasting!

What is driver-based forecasting?

Let’s start from the core. Why do companies want to apply machine learning? This is because traditional forecasting methods, such as time series forecasting, do not take into account impactful demand drivers. In many industries, there are additional drivers that have a far greater effect on demand than for example seasonality.

In machine learning models we incorporate data that relates to these demand drivers, with the aim to learn from their impact in the past in order to reflect this in the forecast. Depending on the industry, internal or external (market) drivers are more relevant to the forecast. In general, internal drivers often have a focused impact on forecasting, such as planning a promotion for a certain range of products at a specific customer or introducing a new product to your biggest customer.

Creating a value-adding machine learning model takes effort and time but can really have a big impact on your forecasting performance. Two of our projects have shown that by applying driver-based forecasting the final forecast (including all manual input from sales and marketing) can be improved by even 12%pt of forecast accuracy and can bring bias within the 2% bandwidth. Another big advantage is that since the most important known demand drivers are already taken into account, only exceptional changes are needed to come to a final forecast.

How to start applying machine learning in forecasting?

As mentioned in the introduction, it is not easy to start applying machine learning in demand forecasting in a proper way. Tooling and knowledge are the biggest hurdles that companies face.

IT departments often have a long-term IT roadmap and need to see how applying machine learning in demand planning would fit in there. On the other side people currently involved in the demand planning process often do not have the technical capabilities to build a small proof-of-concept themselves.

Surprisingly, most companies do not think that the quality and availability of their data should be a roadblock. Let’s take the example of a company, with access to suitable tooling, people with the right skill set, and high-quality data available, then still there is another hurdle to overcome and that is the question of what it will bring? Preparing a proper business case is crucial.

Luckily, that is something where EyeOn can be of help. We can build a proof-of-concept in a very short amount of time. Our data scientists build and apply machine learning models in a very pragmatic and robust way. The proof-of-concept offers you:

• Quick insights into the value of demand drivers on the forecastability of your business
• The maximum forecast performance that can be reached by using driver-based forecasting
• A quantified improvement potential versus current forecasting methods (in efficiency and effectiveness)

The proof-of-concept shows great potential and you would like to grasp the benefits immediately? Our Planning Services team is ready to build and operate your driver-based forecasting process from the start. In the meantime you can work on getting tooling and knowledge to the required level, so we can transfer the operation back to your internal organization.

In this video we further explain the EyeOn driver-based forecasting proof-of-concept:

We are here for you!

Do you want to improve your forecast performance? At EyeOn, we’ve developed the Fast Forecast Scan: a quick tool that provides you with rapid insights into the demand characteristics and forecastability of your business. As a first step, we perform a thorough deep dive into your data and provide actionable data quality insights. With improved data quality, the Fast Forecast Scan provides you, within a few days, with data-backed insights on the highest possible forecast accuracy that can be reached and identifies the main opportunities for improvement. Witness the transformative power of the Fast Scan through our on-demand demo.

You prefer to speak to one of our specialists? Please reach out to us and see how driver-based forecasting can be of added value to your organization: Contact Erik de Vos, Willem Gerbecks, or André Vriens!

The post How to make a next step in forecasting with machine learning appeared first on EyeOn.

Moving holidays are holidays that occur each year, but where the exact timing shifts from the perspective of the Gregorian calendar system. Chinese New Year (CNY) is an example of a moving holiday, it is based on the lunar calendar. Chinese New Year most often falls in February but can also occur in January. Since the date of Chinese New Year changes from year to year, the effect of this holiday can impact sales in multiple months. It is often the case that production accelerates some time before the start of Chinese New Year, almost completely stops during the holidays, and finally rises to the regular level after the holidays. In these cases, the effect of the holiday is not confined to the seasonal component of the time series since the seasonality rhythm (based on the lunar calendar) is not in line with the demand forecast rhythm (based on the Gregorian calendar). This often leads to a significant decrease in the performance of the statistical forecast (i.e. lower forecasting accuracy and higher bias) for the months affected by the holiday.

Combining statistical forecasting with machine learning for increased forecasting accuracy

Conventional statistical models (e.g. moving average and exponential smoothing) are widely used within the industry to predict demand. Often with good reason, since these models usually perform reasonably well and they are intuitive and easy to interpret for planners. We have seen, that statistical models (even if we add a seasonal component) are not able to model the complex effects of moving holidays. Based on these observations, we designed a new approach using machine learning to predict ‘uplift’ factors that are used to scale the baseline forecasts. We give a detailed description on how EyeOn addresses this challenge in this blog post: How to make better demand forecasts for Chinese New Year using machine learning

Results case study previous blog

Going beyond Chinese New Year

The results of modeling Chinese New Year for a large multinational are very promising. In this case study, we find enormous improvements in forecast bias. For ten different regions, we achieve an average absolute bias reduction of 68 percent points. We don’t stop there, however! The next step is to extend the logic on two fronts: (1) automatic detection of significant events (rather than manual) and (2) applying the logic to a wider variety of moving holidays in addition to Chinese New Year.

First, we implement automated event detection to determine whether an event significantly impacts a time series in a certain period. If so, we use our machine learning model to apply a correction to the baseline forecast. If not, we just use our statistical baseline forecast. The event detection consists of three statistical tests, which determine if the sales during an event are significantly impacted. The identified events are corrected by predicting uplift factors for the event periods. Automatic event detection results in less manual intervention from the forecaster and provides statistical proof that an event has an impact on the time series. Making you less dependent on the forecaster’s ability to interpret and their experience of whether a correction should be applied.

Second, besides the uplift factors, the event sales are corrected to improve the forecast performance in the periods after the event. We do this correction to avoid the sales disruption during the event impacting the statistical baseline forecast. In the case of Chinese New Year, the correction is used to make sure that the statistical forecast level will not be too low after having three negatively impacted sales months.

The extension of this logic makes it possible to better forecast moving holidays and to rely on statistics to determine which events should be corrected. Making it applicable for more customers and events, resulting in improved forecasts during moving holidays. We applied the new extended logic to two other events besides Chinese New Year in a new case study. Comparing the results of the developed method to the results of conventional time series models (e.g. moving average/simple exponential smoothing). In this case, we find an average increase of 13.1 percentage points in forecasting accuracy and a reduction of 18.7 percentage points in the bias for the affected months. Note that for the non-impacted months, the two forecasts are identical.

Challenge accepted, challenge completed: increased forecasting accuracy

Moving holidays can have a significant impact on demand. Moreover, modeling this effect can be challenging since the effect of the event impacts a different period (e.g. week or month) each year. We accepted this challenge and developed a new approach where the conventional time series forecast is complemented with a machine learning algorithm that models the effect of moving holidays. This approach is now available for a large variety of events and customers.

Eager to learn more about this topic or curious if this method could also be valuable for your organization? Please reach out to us!

The post Increased forecasting accuracy during moving holidays appeared first on EyeOn.

Curious about the potential of driver-based forecasting for your business? Get in contact with us for a proof-of-concept!

The post Video: Discover the benefits of driver-based forecasting appeared first on EyeOn.

EyeOn offers a framework to leverage the full potential of your demand planners. In this webinar we presented how our framework can guide your demand planners.

Watch the webinar recording below to learn more, or contact us directly!

The post Demand planning 2.0 – Take control of your role! appeared first on EyeOn.