AI in Retail: Applications in the Industry 4.0 Market

AI in Retail: Applications in the Industry 4.0 Market

Currently, we experience the technological advances present in the industries, directly impacting the processes of construction, development, and delivery of products to the consumer. Market competitiveness increasingly focuses on technological and digital pillars, thus making the automation and digitization of processes more recurrent in companies. The impact generated by Industry 4.0 has created a horizon of opportunities for the retail market to get ahead and compete at higher levels with competitors, seeking higher rankings in terms of delivery, quality, efficiency, and effectiveness in the processes until the product reaches the consumer.

AI in retail

In addition, advances in Artificial Intelligence, Machine Learning, and IoT (Internet of Things) provide new horizons for the various branches of the retail industry. The automation of storage processes, route monitoring, material storage strategies, demand forecasting, and customer satisfaction are examples of procedures adopted through these technologies to obtain better results in the market.

We are living in the data age. Being prepared for it, and orienting internal processes to data, will enable companies to immerse themselves in this ocean of opportunities, thus resulting in cost reductions based on analysis of losses and waste, and in more sustainability, competitiveness, and approval in the market.

Supply chain and S&OP

Among the different operational areas of the industry, artificial intelligence stands out strongly in the supply chain, promoting more automation in production processes. Tracing a procedural path of operations with AI, it is clear that the cycle ranges from the implementation of intelligent technologies in sales and operations (S&OP) processes, thus intensifying the analysis for better sales strategies with the help of the marketing area, to better formats to operate and the automation of exhausting and repetitive jobs.

Through the advancement of the Internet of Things, it becomes more efficient to capture data from different stages of production. Obtaining data from the first task to delivery to the final consumer is no longer a problem, with the possibility of extracting production data, for example, robots implemented for storage, applications for drivers, and products connected to the internet, among other ways of collecting data with IoT. Another important point is the bridge between stakeholders through advanced data analysis with Machine Learning and AI, aiming to filter raw material suppliers and final suppliers that are more aligned with the company’s interests and also to obtain a lower loss in the processes. procurement of materials and delivery to the consumer.

AI, ML, and IoT technologies also influence revenue generation, increasing profits and better results with customer and supplier relationship management. An example of this is intelligent dynamic pricing, which uses artificial intelligence and adopts market and consumer-based strategies to determine the best price (not necessarily the highest, but the most appropriate price to compete in the market), aiming at increasing revenue.

Demand Forecasting 

Regarding demand forecasting processes, the implementation of AI and ML produce an assertiveness of around 90%, generating impact and improvements in demand forecasts based on advanced analysis of different data, such as weather conditions, the economic situation of the market, available quantities, consumer desire, and consumption predictability. In addition, advanced analytics and intelligent models that have continuous learning through greater data collection and time provide predictive actions in real-time, helping decisions in a way assisted by professionals. This reduces the failures and risks in operations with decision-making and can change them in case of negative predictions that can generate several impacts.

Furthermore, in the area of ​​dairy and perishable products, AI has great strength, as strategies for goods with short dates and more fragile logistics need to be much sharper. This contribution is supported by collecting data, and information and creating predictive demand models that deliver better strategies for storing products, defining the best routes, reducing fuel waste, and forecasting geolocation in the case of products with greater demand. so as not to keep them in distant stocks, thus facilitating the preservation of the products until their final delivery.

Big Data

This is a term that has been gaining great proportion and space in the context of industry 4.0, representing the large mass of data, intensive collection, and importance of artificial intelligence and machine learning to handle this information that can add a lot of value to companies. Represented by the thousands of data produced by the different stages and experiences of the market, big data includes purchase data, online browsing of consumers, media and marketing data, and customer satisfaction with the service and/or product, among other diverse information.

The process of collecting and storing data is complex and analyzing thousands of data becomes an impossible human task. Thus, AI and intelligent models based on machine learning go hand in hand with big data to integrate external market and internal company data in a way that makes forecasting and planning of demand, greater revenue, profit, reduction of waste, and sustainability.

Logistics 4.0 

It is clear the advances that industry 4.0 has been allocating. For example various automation in production processes, digitalization of products for testing improvements, speed of information, and implementation of results.

With industry 4.0 comes logistics 4.0, aimed at optimizing the loading and unloading processes of goods. Automation and use of AI in various stages of logistics, such as the organization of products in warehouses made by robots that, through AI, strategically and hierarchically organize products to facilitate and increase the speed of cargo operations.

In addition, it is possible to generate forecasts of events on highways, such as works that interrupt routes, using AI and real-time data analysis. This allows the adoption of a better route in the present time, without relying on the historical past and wasting resources, also resulting in customer satisfaction and speed of delivery. Taking advantage of routes, inappropriate and unnecessary use of vehicles, higher gas emissions, and high fuel and maintenance costs are problems interrupted by logistics 4.0 directives, aiming at more assertiveness, intelligence, sustainability, greater revenue, and consumer and supplier satisfaction.

AI in Retail – Final Thoughts

Implementing AI and machine learning through intelligent models is not an easy and instantaneous task. However, the result of all the preparation and construction of these technologies directed to the specifics of the business will result in several benefits.

The power of AI provides intelligent market insight, demand forecasting with higher hit rates, reduced product loss due to expiration or warehouse saturation, and precision in price adjustments supported by different variables that can influence revenue variation. In addition, through advanced data analysis, it is possible to filter suppliers looking for those that deliver the most results and are more aligned with the company’s values.

These positive points are in line with the use of AI to obtain better sustainable results, aiming at the use of routes, continuous delivery, analysis of better routes, reduction in costs, and gas emissions.

The access generated by advanced analytics and AI of the entire supply chain and operations of the company results in great predictability of risks or failures in the initial stages, preparation, and delivery to the end customer. This power of predictability and intelligent strategies consolidates the idea of ​​risk management in a real-time, drastic reduction of failures and waste, and unified control of the stages of sales, operations, production, and delivery of goods. In short, smarter and more sustainable companies have never been so close to being consolidated. The way forward only depends on preparation and organization for greater intelligence and predictability.

What is Aquarela Advanced Analytics?

Aquarela Analytics is Brazilian pioneering company and reference in the application of Artificial Intelligence in industry and large companies. With the Vortx platform and DCIM methodology, it serves important global customers such as Embraer (aerospace & defence), Scania and Randon Group (automotive), Solar Br Coca-Cola (beverages), Hospital das Clínicas (healthcare), NTS-Brasil (oil & gas), Votorantim Energia (energy), among others.

Stay tuned following Aquarela’s Linkedin!


5 real data challenges in the food sector

5 real data challenges in the food sector

The food sector and food security are a global concern and Brazil is one of the main countries responsible for the world demand for food (Estadão). In this sense, what are the main challenges related to data management to optimize Brazil’s operational efficiency in the food/agribusiness sector, which today represents 21% of Brazil’s GDP?

This article addresses the issue with the bias of Aquarela’s experience in Advanced Analytics and Artificial Intelligence projects carried out in large operations in Brazil. The risk of a lack of information is as relevant as its excess and lack of analysis, which can impact the efficiency of the sector’s logistics chain as a whole.

Below, we have elaborated on some of these main risks.

Characterization of the food sector

The food sector is quite varied due to the large extension of the production chain, which ranges from agricultural inputs, industrialization, transport logistics to commercialization in consumer markets and finally the final consumer.

As fundamental characteristics, the food sector is directly linked to factors that can have great variability and little control, such as: 

  • Climate (temperature, water volume, luminosity and others);
  • Economic factors such as currency fluctuations;
  • Infrastructure;
  • Domestic/external market demand.

In addition to these factors, below we list some related to data management. We also show how they, if well organized, can help mitigate the effects of uncontrollable variables in the food supply chain.

01 – Incompleteness of information

The supply chain is quite large. This makes the data complex and difficult to interpret due to the different phases of each process, culture and region. In addition, it causes many important planning decisions to take place with very limited information and high risk. In other words, decisions are made without a vision of the complete scenario of the chain, largely following the manager’s intuition.

The lack of quality information is a big risk. If data is lacking today, imagine what the scenario was like 10 or 20 years ago.

In recent years, the industry and retail have shown great advances in their computerization processes with various traceability solutions. With the evolution of Industry 4.0 technologies (IOT and 5G) in the coming years, it is likely that the food market, from the agricultural and industrial sector to the commercial sector, will hold more complete information for decision making than what is currently available today.

02 – Data from multiple sources

If data is becoming more and more present with the development of informatization and communication, then the next problem is trying to analyze data from multiple and disconnected sources.

Different data is often stored on different systems, thus leading to incomplete or inaccurate analyses. Combining data manually to form datasets (what are datasets?) for analysis is quite heavy and time-consuming work and can limit insights into the reality of operations.

What is sought is the construction of Data Lakes adherent to the type of management to democratize access to data by market professionals, thus optimizing their activities with increasingly powerful analytics solutions. This not only frees up time spent accessing multiple sources, it also allows for cross-comparisons and ensures that the data is complete.

03 – Low quality data

Having incorrect data can be just as or more harmful than not having it. Nothing is more harmful to data analysis than inaccurate data, especially if the idea is to use data science and machine learning practices. Without a good input, the output will be unreliable.

One of the main causes of inaccurate data is manual errors made during data entry, especially when information is collected manually. Another problem is asymmetric data: when information from one system does not reflect changes made to another system and thus makes it out of date.

Analytics strategic planning projects seek to mitigate and/or eliminate these problems. This happens from systematic processes of data dictionarization, survey of processes, functions, and so on.

04 – Lack of data talents

Some organizations and companies, in general, are not able to achieve better levels of efficiency in operations, as they suffer from a lack of talent in the area of ​​data analysis. In other words, even if the company has consistent technologies and data, the manpower to execute the analysis and action plans still counts a lot at the end of the day.

This challenge can be mitigated in three ways:

  • Develop an analytical technology stack that is always up-to-date and adherent to the business and with up-to-date training materials.
  • Add analytical skills to the hiring process. In addition, invest in the constant training of the team on new data technologies related to the technological stack of the operation.
  • Use analytics outsourcing to accelerate the process. In this article, for example, we list the main aspects to be considered when choosing a good supplier.

05 – Customization of values ​​and product characteristics in the food sector

Although, according to Embrapa, about 75% of the entire world food sector is based on just 12 types of plants and 5 types of animals, there are thousands of different products, marketed in multiple ways, prices and deadlines in the final consumer market.

Just as an example, in the area of ​​animal protein, the process of marketing cattle meat requires investments, infrastructure, deadlines and processes that are quite different from what would be for the production of pork or even chicken.

Since the processes are different, the data generated by the production chain also becomes different, requiring customizations in information systems and databases. As a consequence, there are changes in models of:

The recommendation is to parameterize the systems based on the most common classifications in the market and focus on the most important products from a strategic point of view (contribution margin, volume or sales price).

5 real data challenges in the food sector – Final thoughts

In this article, we have collected some relevant points about the real challenges of data in the area of ​​food, a sector in which  Brazil stands out as one of the main global players.

It is a complex area with several risk factors and great opportunities for optimization with the increasingly intensive use of data. Previously, we wrote an article related to data strategies for energy trading and which in part has the same challenges related to decision making in the food sector.

We, at Aquarela Analytics, constantly work with these challenges of making the complex  things simple and with good risk mitigation. So if you have any questions, get in touch with us!

What is Aquarela Advanced Analytics?

Aquarela Analytics is Brazilian pioneering company and reference in the application of Artificial Intelligence in industry and large companies. With the Vortx platform and DCIM methodology, it serves important global customers such as Embraer (aerospace & defence), Scania and Randon Group (automotive), Solar Br Coca-Cola (beverages), Hospital das Clínicas (healthcare), NTS-Brasil (oil & gas), Votorantim Energia (energy), among others.

Stay tuned following Aquarela’s Linkedin!


Dynamic Pricing in the Private Education Business

Dynamic Pricing in the Private Education Business


The pricing process is a strategic business activity that requires continuous analysis and information sharing between departments in order to be assertive and financially beneficial. Over the years, it has been getting optimized through increasingly intelligent, dynamic processes powered by technology.

The intelligent dynamic pricing strategy is used to generate prices based on the analysis of information captured in the market (competition) and even on probabilistic values ​​generated by the use of Artificial Intelligence (AI). This innovative approach has gained prominence among organizations from different sectors, starting to be adopted by the main companies in the digital world or in the process of digital transformation, offering agile market adaptability, competitiveness, and profit maximization.

Based on that, today in our blog we are going to present dynamic pricing in private education, as it is one of the most important and challenging decisions in the sector.

How to value the brand and, at the same time, consider the investment possibilities of your students? How to price correctly in a market where scholarships are often offered? And yet, how to define the ideal tuition fee/scholarship for each student and at the same time ensure uniformity in the application of corporate pricing policies?

To answer these questions, we have prepared a material (which you can download for free), presenting our success story related to advanced pricing in the private education segment.

The case study addresses our client’s business problem, who had been immersed in a complex non-automated pricing system, with extensive rules to define the ideal tuition fee/scholarship for each student; the journey, marked by the structuring of the pricing process; and the results generated for the educational institution (what the client gained).

Dynamic Pricing in the Private Education Business: Conclusion

The dynamic pricing strategy has been gaining prominence in the education sector. However, its implementation process is complex, requiring artificial intelligence as well as qualified, market-recognized data analytics providers.

Also read – How to choose the best AI and data analytics provider?

Do you have any doubts about our success story of dynamic pricing in private education? Leave your comment.

Learn more about Smart Pricing, Aquarela Tactics module, and talk to our experts.

What is Aquarela Advanced Analytics?

Aquarela Analytics is Brazilian pioneering company and reference in the application of Artificial Intelligence in industry and large companies. With the Vortx platform and DCIM methodology, it serves important global customers such as Embraer (aerospace & defence), Scania and Randon Group (automotive), Solar Br Coca-Cola (beverages), Hospital das Clínicas (healthcare), NTS-Brasil (oil & gas), Votorantim Energia (energy), among others.

Stay tuned following Aquarela’s Linkedin!


Dynamic Pricing: definition and benefits

Dynamic Pricing: definition and benefits

Dynamic pricing, also called smart pricing, price optimization, pricing customization, pricing recommendation, or strategic pricing, is a relevant point in strategic planning of companies which has been increasingly calling for special attention from CEOs, CIOs, sales management and revenue management.

According to Forbes, the main companies in the digital world who seek to stand out in a competitive market use dynamic or smart pricing. This kind of strategy can be applied in different sectors, helping companies achieve the quality and pragmatic marketing necessary for success.

In this article, we are going to show some benefits that dynamic pricing brings.

What is dynamic pricing ?

Dynamic pricing is a strategy used to establish flexible market prices for products and services offered by companies, taking into account some variables such as balance between supply and demand, seasonality and competitive strategy. This strategy allows companies to adapt to the market quickly and competitively.


For a better understanding of the advantages of investing in dynamic pricing we show you some benefits that this strategy can provide for your business.

Profit Maximization

Using dynamic pricing, it is possible to track the value of one specific product more quickly, and simultaneously track competitor price fluctuations. This offers a vantage point that allows you to follow trends in order to increase your profit margin, without fearing the loss of market space. When a market trend results in the lowering of prices, it’s possible to prepare to increase the number of sale conversions then profit by volume as result.

Understanding the consumer behavior 

Another dynamic pricing benefit is the fact that it assists in understanding and addressing key consumer behavior points, for example, the types of promotions that work best and the best-selling products in a given period, among other things.

Benchmarking Automatization

In addition to allowing more competitive prices compared to the competition, dynamic pricing is a way to quickly keep up with the competition, allowing the company to prepare against threats or seize opportunities. It can also help in observing competitors to create attractive differentiators for your company.

Dynamic Pricing Implementation 

Based on sales indicators and pricing data, it is possible to create a tool to review prices at the desired frequency, based on the company’s rules. The recommendation is that the data be structured through a Data Lake. 

The usage of Advanced Analytics techniques and Artificial Intelligence (AI) algorithms foster improvements in pricing systems. By analyzing past data,  it’s possible to look for successful pricing patterns,  or to produce complex models for AIs. This allows for the development of tools driven by strategic goals ​​that can define prices based on present or future factors. In other words, AI can help organizations reach the point where they’d like to be.

Want to know more about how to implement dynamic pricing in your company? Click here.

Conclusions and recommendations

Dynamic pricing is a huge opportunity for your company’s growth, when you take advantage of all that it offers, it allows for superior planning while avoiding loss in conversions.

The decision to make prices dynamic is not simple, so it is important to choose a suitable data analytics supplier. Some aspects to be analyzed when looking for a supplier for the pricing solution can be found in this article: How to choose the best AI and Data Analytics provider?

Do you like our content? rite a comment and share on your social media.

What is Aquarela Advanced Analytics?

Aquarela Analytics is Brazilian pioneering company and reference in the application of Artificial Intelligence in industry and large companies. With the Vortx platform and DCIM methodology, it serves important global customers such as Embraer (aerospace & defence), Scania and Randon Group (automotive), Solar Br Coca-Cola (beverages), Hospital das Clínicas (healthcare), NTS-Brasil (oil & gas), Votorantim Energia (energy), among others.

Stay tuned following Aquarela’s Linkedin!


Demand Forecasting: all you need to know about this subject

Demand Forecasting: all you need to know about this subject

In business or in our personal lives, we are used to making predictions and estimates about future events. For example, the estimated time taken to arrive at a friend’s house or the amount of food needed for a dinner for several people. Decisions made based on  a forecast can make us arrive at the right time for a meeting or buy the right amount of food or not. With that in mind, we at Aquarela have prepared this article on demand forecasting, covering its concepts, importance for business and evolution over the years.

What is Demand Forecasting?

Demand forecasting, as the name already tells, is the process for obtaining estimates of a future demand of products and services. This is done with existing data that’s been collected and stored. Forecasts can be generated through mathematical models, which use historical data; by qualitative methods, such as management experience or expert opinion; or even a combination of both.Some of the questions we seek to answer when making such predictions are “when”, “where”, and “how much”. It is important to note that predictions are not targets, but rather estimates of what will happen. Forecasting results aid businesses in planning their decision making. Therefore, a good demand forecasting process needs to produce good estimates.

How important is Demand Forecasting?

Demand Forecasting is an important activity that influences companies from different segments, such as: retail, consumer goods, pharmaceutical industry, automotive electronics, heavy machinery among others.

Demand forecasting is used for business planning, as every plan involves estimates about this type of forecast. Thus, the predictions are of importance, as they enable managers to plan more assertive activities toward the strategic goals of the business. It is also useful for the tactical and strategic process of companies. Managers and decision makers utilize demand forecasts on their daily activities. These predictions can also be used as inputs for the sales and marketing teams to create insights into demand generation and organize their actions.

Businesses, in general, need to make forecasts about their products to prepare an effective plan in both short and long term. With that in mind, demand forecasts are crucial for companies, as they affect inventory planning, logistical planning, production planning, cash flow planning, hiring decisions, purchasing decisions, among others. Poor or absent forecasting can lead to bad decisions. So, without good demand forecasting, businesses  would be poorly preparing themselves for the future events.


Better forecasts can result in better service levels, customer retention, cost savings, waste reduction (excess inventory and unsold products). In addition, the need for emergency production to meet unexpected demand is relieved, because, with predictions, companies can plan and make decisions that allow a better response. 

Not being prepared for the demand can cause incalculable losses, such as the reduction in market share. For example, when a customer needs a product, but can not find the desired brand, he can often find a replacement with a competing brand. Due to this, companies can no longer be reactive to meet demands for their products. Furthermore, the exponential growth of supply chains belonging to certain companies was only possible due to demand forecasting improvements. They invest so that they can plan better, because they know that it generates a financial return. It is important to know what is happening at points of sales to obtain a more assertive demand forecast.

How are forecasts horizons classified?

The demand forecasts can be sorted by what are called “future horizon” in the time it applies. The three classification categories in relation to the time horizon are:

  • Short-Range Forecast: usually less than three months, but can be up to one year. Used for purchase plans, production scheduling, predictions of labor and production levels.
  • Medium-Range Forecast: Usually three months to one year, however can be from one year to three years. Useful for sales planning, operation plan analysis, production planning and budget planning.
  • Long-Range Forecast: Three years or more. This kind of forecasting can be applied to planning new products, business expansion, as well as research and development.

Short-range forecasts tend to be more assertive than long-range forecasts.

What are the main categories of demand forecasting methods?

The methods of demand forecasting fall into three basic categories, which are: 

  1. Qualitatives: methods that work based on opinions data of managers, experts, sales staff, and customers questionnaires to estimate the quantitative value of demand.
  2. Time series: statistical approach that relies on historical data to forecast future demand. Through these methods, there is a recognition of seasonal trends and patterns.
  3. Causal models: quantitative method of demand forecasting that uses historical data together with independent variables, such as economic conditions, competitors’ actions and promotional campaigns.

Regardless of the category of demand, a prerequisite is that there is a pattern or relationship that can be identified and modelled.

Regardless of the category of demand, a prerequisite is that there is a pattern or relationship that can be identified and modelled.

Evolution of demand forecast throughout history

Organizations began creating demand forecasting departments and functions in the late 1980s. However, this subject was already explored in previous decades in academia and by segments of products that need different parts to be assembled. The beginning of demand forecasting in companies, for the most cases, consisted of simple statistical models , such as moving averages, simple exponential smoothing and judgment by instinct (“gut feelings”).

Technological advances in data storage and processing (Big Data) have positioned demand forecasting as one of the main actors of value generation for Supply Chain. As a result of these advances, demand forecasting is getting better and better.

Conclusions and Recommendations 

Forecasting future demands is a challenge that companies have to face in order to be able to make decisions that allow them to compete by generating better supply chain results. Demand forecasting is an essential activity for business planning, as it results in several benefits, such as: reduced waste, better allocation of resources, increased sales and revenue. This way, it helps organizations to be in the right place, at the right time, with the right product.

Also read: AI for demand forecasting in the food industry

Keep improving your companies’ demand forecasting process and technologies to keep pace with this ever-changing world. Aquarela Tactics has a demand forecasting module that can help you with this journey to know in advance what to expect from the consumer market and optimize your decision making.

What is Aquarela Advanced Analytics?

Aquarela Analytics is Brazilian pioneering company and reference in the application of Artificial Intelligence in industry and large companies. With the Vortx platform and DCIM methodology, it serves important global customers such as Embraer (aerospace & defence), Scania and Randon Group (automotive), Solar Br Coca-Cola (beverages), Hospital das Clínicas (healthcare), NTS-Brasil (oil & gas), Votorantim Energia (energy), among others.

Stay tuned following Aquarela’s Linkedin!


AI for demand forecasting in the food industry

AI for demand forecasting in the food industry

The concept of a balance point between supply and demand is used to explain various situations in our daily lives, from bread in the neighborhood bakery, which can be sold at the equilibrium price, which equals the quantities desired by buyers and sellers, to the negotiation of securities of companies in the stock market.

On the supply side, a definition of the correct price to be practiced and mainly the quantity are common issues in the planning and execution of the strategy of several companies.

In this context, how are technological innovations in the data area establishing themselves in the food sector?

The construction of the demand forecast

The projection of demand is often built through historical sales data, growth prospects for the sector or even targets set to engage sales of a certain product.

When considering only these means of forecasting, without considering the specific growth of each SKU (Stock Keeping Unit), companies can fall into the traps of subjectivity or generalism.

The expansion of a sector does not result in a growth of the same magnitude for the entire product mix. For example, does a projected annual growth of 6% for the food sector necessarily imply equivalent growth for the noble meat segment?

Possibly not, as this market niche may be more resilient or sensitive than the food sector, or it may even suffer from recent changes in consumer habits.

Impacts of Demand Forecasting Errors

For companies, mainly large ones with economies of scale and geographic capillarity, an error in the forecast of demand can cause several consequences, such as:

  • Stock break;
  • Perishable waste (What is FIFO?);
  • Drop in production;
  • Idle stock (slow moving)
  • Pricing errors

Adversities like these directly impact the companies’ final results, as they result in loss of market share, increase in costs or low optimization in the dilution of fixed costs, growth in the loss of perishable products, frustration of employees in relation to the goals and mainly break in the confidence of recurring customers who depend on supply for their operations.

The demand forecast in the food sector

The food industry is situated in a context of highly perishable products with the following characteristics:

  • High inventory turnover;
  • Parallel supply in different locations;
  • Large number of Skus, points of production and points of sale;
  • Verticalized supply chain;
  • Non-linearity in data patterns;
  • Seasonality.

These characteristics make the sector a business niche that is more sensitive to deviations in demand forecast and adjacent planning.

Supply chain opportunity

As an alternative to the traditional demand forecast format, there are opportunities to use market and AI data to assist managers in the S&OP (Sales & Operations Planning) process, as well as in the S&OE (Sales and Operations Execution) process.

During the S&OP process, demand forecasting supported by AI facilitates the work of the marketing and sales areas, as well as reducing uncertainty and increasing predictability for the supply chain areas.

In the S&OE process, AI can be used to identify new opportunities and to correct deviations from what was planned.

In addition to the technical attributes that AI can add to the process, the data base reduces points of conflict between teams, reduces historical disputes between preferences for SKUs and makes the process more transparent between areas.

Previously, in our blog, we addressed the challenges of forecasting demand in our view (pt. 1 in portuguese). In the articles, we cite the differentials of the predictive approach in relation to demand, taking into account factors such as seasonality, geographic / regional preferences and changes in consumer behavior.

We understand that the need for a predictive approach through data, mainly external to the company, is increasingly latent.

The role of machine learning in the food sector

The use of AI through machine learning techniques associated with a coherent technological stack of analytics (What is a technological stack?) Provides greater information speed, data organization with different granularities (region, state, city and neighborhood), adjustments seasonality, exploration of opportunities and decision making in real time.

In the case of the food sector, the greatest accuracy in forecasting demand means:

  • Inventory optimization among Distribution Centers (CDs);
  • Reduction of idle stocks;
  • Decrease in disruptions that cause loss of market share due to substitute products;
  • Direct reduction in losses with perishability (FIFO).

The great technical and conceptual challenge faced by data scientists (The profile of data scientists in the view of Aquarela), however, is the modeling of analysis datasets (what are datasets?) That will serve for the proper training of machines.

Please note that:

“Performing machine training with data from the past alone will cause the machines to replicate the same mistakes and successes of the past, especially in terms of pricing, so the goal should be to create hybrid models that help AI replicate with more intensity and emphasis the desired behaviors of the management strategy “.

In the case of Aquarela Analytics, the demand forecast module of Aquarela Tactics makes it possible to obtain forecasts integrated into corporate systems and management strategies. It was created based on real national-wide retail data and algorithms designed to meet specific demands in the areas of marketing, sales, supply chain, operations and planning (S&OP and S&OE).

Conclusions and recommendations

In this article, we present some key characteristics of the operation of demand forecasting in the food sector. We also comment, based on our experiences, on the role of structuring analytics and AI in forecasting demand. Both are prominent and challenging themes for managers, mathematicians and data scientists.

Technological innovations in forecasting, especially with the use of Artificial Intelligence algorithms, are increasingly present in the operation of companies and their benefits are increasingly evident in industry publications.

In addition to avoiding negative points of underestimating demand, the predictive approach, when done well, makes it possible to gain market share in current products and a great competitive advantage in forecasting opportunities in other niches before competitors.

What is Aquarela Advanced Analytics?

Aquarela Analytics is Brazilian pioneering company and reference in the application of Artificial Intelligence in industry and large companies. With the Vortx platform and DCIM methodology, it serves important global customers such as Embraer (aerospace & defence), Scania and Randon Group (automotive), Solar Br Coca-Cola (beverages), Hospital das Clínicas (healthcare), NTS-Brasil (oil & gas), Votorantim Energia (energy), among others.

Stay tuned following Aquarela’s Linkedin!