SodaStream, an Israeli manufacturer of fizzy drink devices, gained visibility in the U.S. and Europe as a healthy and environment friendly alternative to carbonated giants like Coca Cola. But soon after relocating from a controversial site in the occupied West Bank to a new facility in southern Israel, executives realised that the company is facing a new challenge: streamlining operations in order to stay competitive with low-cost manufacturer rivals from China while quenching a fast-growing thirst for its bubbly beverages. To rein in costs and make SodaStream's four manufacturing lines more efficient, executives decided to automate assembly lines with robots, computerise production, and connect all manufacturing processes under one control system. The multi-year project was aimed at boosting output to keep pace with 30 percent yearly sales surges, while utilising artificial intelligence, machine learning and cloud computing to get a better handle on optimising production. "We continued to grow rapidly and were packed with endless employees. The dining room was full. The production side was full. We knew that we wouldn't be able to allow ourselves to keep operating the same way… whether in terms of space, efficiency, or in terms of costs," said Kfir Suissa, chief operation officer at SodaStream, which was acquired by PepsiCo in 2018 for US$3.2 billion.
Disruptive technological solutions are transforming the Food & Beverages industry from the way food is manufactured, transported and marketed, through the agility of internal processes to the way products are communicated to consumers. Data Analytics and Artificial Intelligence bring benefits to companies at the strategic level, through decisions based on reliable data, more flexible and automated production, and better adaptation to changing markets. In this eBook, discover how this industry can be more disruptive in its offer, more agile and intelligent and with a deep knowledge of an increasingly demanding and informed consumer's preferences and trends.
Artificial intelligence (AI) traces its history to the 1950s with Alan Turing and the Dartmouth Summer Research Project, and only in recent years has it emerged as a technology poised to impact society on par with the industrial revolution in terms of eliminating whole job segments such as truck drivers. The big question is will it create as many new jobs as it eliminates.
Digital disruption is affecting nearly every industry, from financial services to healthcare -- and the food and beverage sector is no exception. Historically, flavor profiles, trends and new food products have largely been attributed to chefs and product developers, and it would take months or years before an idea could be translated into a product and introduced to the market. In more recent years, however, the answer to the next big food or flavor trend has had less to do with humans and more with the power of big data and artificial intelligence (AI), which learns and mimics human behavior by collecting and analyzing millions of data sets concurrently. So how does harnessing technology translate into the next flavor or trend? As an example, spice company McCormick partnered with IBM in 2019 to leverage AI to predict new flavor combinations.
The use of AI-driven processes to increase efficiency in the F&B market is no longer an anomaly. A host of breweries and distilleries have incorporated the technology to not only develop flavour profiles faster, but also for other functions, including packaging, marketing, as well as to ensure they meet all food-safety regulations. Although the intention is not to find a replacement for the brewmaster/distiller, it becomes a thrilling learning experiment that equips them with multiple data points that could help them come up with innovative ideas. The company claims to be the world's first to use AI algorithms and machine learning to create innovative beers that adapt to users' taste preferences. Based on customer feedback, the recipe for their brews goes through multiple iterations to generate various combinations.
Choosing the technique that is the best at forecasting your data, is a problem that arises in any forecasting application. Decades of research have resulted into an enormous amount of forecasting methods that stem from statistics, econometrics and machine learning (ML), which leads to a very difficult and elaborate choice to make in any forecasting exercise. This paper aims to facilitate this process for high-level tactical sales forecasts by comparing a large array of techniques for 35 times series that consist of both industry data from the Coca-Cola Company and publicly available datasets. However, instead of solely focusing on the accuracy of the resulting forecasts, this paper introduces a novel and completely automated profit-driven approach that takes into account the expected profit that a technique can create during both the model building and evaluation process. The expected profit function that is used for this purpose, is easy to understand and adaptable to any situation by combining forecasting accuracy with business expertise. Furthermore, we examine the added value of ML techniques, the inclusion of external factors and the use of seasonal models in order to ascertain which type of model works best in tactical sales forecasting. Our findings show that simple seasonal time series models consistently outperform other methodologies and that the profit-driven approach can lead to selecting a different forecasting model.
Consumers have always demanded innovation from the retail industry. Shopping habits and product demands are constantly evolving, and retailers invest a significant amount of capital to monitor trends and cater to fluctuating behaviors. Recently, advancing technology has quickened the pace of change and made it even harder to win consumer attention in an increasingly crowded marketplace. More than ever, success requires financial and managerial flexibility and adaptiveness--areas where private equity can play a vital role. Below, read my thoughts on three key areas where our industry is partnering with retailers to help them keep ahead in the fast-changing sector.
We use distributionally-robust optimization for machine learning to mitigate the effect of data poisoning attacks. We provide performance guarantees for the trained model on the original data (not including the poison records) by training the model for the worst-case distribution on a neighbourhood around the empirical distribution (extracted from the training dataset corrupted by a poisoning attack) defined using the Wasserstein distance. We relax the distributionally-robust machine learning problem by finding an upper bound for the worst-case fitness based on the empirical sampled-averaged fitness and the Lipschitz-constant of the fitness function (on the data for given model parameters) as regularizer. For regression models, we prove that this regularizer is equal to the dual norm of the model parameters. We use the Wine Quality dataset, the Boston Housing Market dataset, and the Adult dataset for demonstrating the results of this paper.
It is crucial for the wine industry to have methods like electronic nose systems (E-Noses) for real-time monitoring thresholds of acetic acid in wines, preventing its spoilage or determining its quality. In this paper, we prove that the portable and compact self-developed E-Nose, based on thin film semiconductor (SnO2) sensors and trained with an approach that uses deep Multilayer Perceptron (MLP) neural network, can perform early detection of wine spoilage thresholds in routine tasks of wine quality control. To obtain rapid and online detection, we propose a method of rising-window focused on raw data processing to find an early portion of the sensor signals with the best recognition performance. Our approach was compared with the conventional approach employed in E-Noses for gas recognition that involves feature extraction and selection techniques for preprocessing data, succeeded by a Support Vector Machine (SVM) classifier. The results evidence that is possible to classify three wine spoilage levels in 2.7 seconds after the gas injection point, implying in a methodology 63 times faster than the results obtained with the conventional approach in our experimental setup.
Futurist and artist Syd Mead has passed away at 86 due to complications from lymphoma. Even if you don't know his name, you've probably felt his impact on Hollywood, especially on the science fiction genre. Mead designed Blade Runner's world and technologies by serving as Ridley Scott's concept artist, and he conjured up the lightcycle for Tron, among other fictional vehicles and gadgets. His ideas of the future also helped shape other sci-fi films' universe, including Elysium and Tomorrowland. Mead's background in industrial design may have helped him think up advanced technologies that are still believable.