Electric Vehicles and their Impact on Automotive Warranty Management
The falling costs of the lithium-ion battery pack, coupled with the rising concern of climate change, in addition to policy incentives, rising incomes, and technological advancements have pushed the adoption and sales of electric vehicles (EV) this last decade. The Growth of the EV Segment Bloomberg NEF’s Electric Vehicle Outlook study forecasts that EVs will hit 10% of global passenger vehicle sales by 2025, growing to nearly 28% in 2030 and 58% in 2040. In the US alone, the number of brands offering EV options will grow from 16 now to at least 40 in 2025, with electric vehicles offering consumers a wide range of segments and price points. In addition to climatic awareness, governments are also driving change. US President Biden began his term with an order to electrify the federal fleet of about 650,000 vehicles, investing in over 550,000 public charging points, and taking initiatives to bolster the domestic supply chain for critical technologies and raw materials. The Impact of EVs on the Automotive Industry All of this indicates that the next decade will see a major shakeup in the automotive industry. One example is an estimation by Ford, which believes that the simplification in the assembly of EVs could lead to a 50% reduction in capital investments and a 30% reduction in labor hours, as opposed to internal combustion engine (ICE) manufacturing. Similarly, as the number of electric vehicle consumers grows, other aspects of the automotive industry including automotive design, supply chains and production processes will undergo a transformation as well. As vehicles become more computer-dependent and less combustion engine related, learning gaps are expanding for technicians and even drivers, as they learn to handle what is essentially a new operational vehicle. EV Trends and Future of Warranty Management The automotive industry is already bracing for a shake-up when it comes to repair and parts management. With the surge in Electric Vehicles across the world, we can expect the OEMs to face an increase in the volume of technical warranty requests from their dealers. And because every vehicle is different, there will be a significant shift in how to handle these claims initially. The service and maintenance of an electric vehicle are likely to be highly different from a typical combustion engine, to which the industry has so far been geared. Not only do EVs have fewer mechanical parts, but some components also (such as plugs and sockets, inverters and powerpack coolers), aren’t part of the existing automotive warranty service. How Connected EVs Change the Repair Game There are also several different types of EVs, such as hybrid and connected vehicles, which could further add more complexity to the issue of long-term service and maintenance. IoT devices are also enabling vehicles to stay connected and detect vehicle failures even before they physically reach the service center. That means connected EVs will offer vehicle owners the ability to self-service and distinguish between actual car failure and driver solvable issues. This will impact automotive servicing to a greater extent as manufacturers may need to include ability to educate the driver if vehicles are brought in with no actual failure. There are also likely to be more auto repairs in the field, as the parts get smaller and more computer-like. This could have a significant cost and operational impact on manufacturer warranties as mechanics (with computerized knowledge) travel to the customer rather than the other way around. Smarter Cars, Smarter Warranty Management To manage this transition, OEMs will need to become smarter about their warranty management processes. By using technology to examine warranty claims, OEMs can bring increased efficiency and transparency into a complex process. This will, in turn, enable OEMs to offer superior customer-centric service. The next generation warranty management software solution leverages artificial intelligence and machine learning capabilities in order to identify and understand patterns in warranty claims. Additionally, software solutions such as end-to-end warranty lifecycle management can help OEMs reduce costs in warranty management, increase supplier recovery, and improve aftermarket sales support. Warranty management solutions can also be used to handle increased volumes in claims processing easily. Through machine learning and image recognition, the system can be trained to recognize parts and models automatically, or even detect fraud claims, saving manufacturers a tremendous amount of time and money. Shifting Gears to Stay Ahead The transition to smarter electric vehicles and the potential phasing out of combustion engines is likely to be a game-changer for many. Automotive manufacturers and suppliers are making key investment and technology decisions about the next generation of vehicle and components manufacturing, already. Forward-thinking OEMs will need to tackle their challenges by using technology solutions to enable transparent cooperation with partners to offer sourcing, supply, and maintenance benefits and future proof their business. SOURCES: The future of cars is electric – but how soon is this future? The Auto Industry and EVs: Where We Are and What’s Coming Next, After Years of Crying Wolf? Plugging Into The Future: The Electric Vehicle Market Outlook
Making Warranty Management Profitable for Manufacturers
Traditionally, manufacturers offered warranties to buyers to assure buyers of the quality and longevity of products or services. The manufacturer (or the seller) would typically cover the repair or replacement of the products within a period of time. The industry has progressed a long way from there. With the increasing complexity of production and the supply chain, warranties today include everything from product coverage to customer support and logistics. But today, with distribution and supply chains spanning continents, how do manufacturers stay on top of their warranty information and claims management? What is Warranty Management? IDC Manufacturing Insights defines warranty management as the stages of the warranty process, including registration, claim capture, claims validation, early failure detection, recalls, parts returns, adjudication, extended warranty service, supplier recovery, and reserve optimization. Professional claims management for manufacturers often comprises the paperwork for warranties, checking validity in case of fraudulent claims, and ensuring fast, efficient and cost-effective processing while fulfilling warranty claims. Managing warranties often includes the stakeholders (and their roles) in ensuring customer satisfaction with a product or service. If you are looking for a Warranty Management system that can help you reduce backlog, and minimize claims processing time and paperwork, check out Tavant’s solutions. The Impact of Digitization on Warranty Management With increased digitization, more manufacturers are turning to technology solutions to optimize the warranty management process. While manufacturers have traditionally viewed warranty management as a necessary evil, studies now prove that the right technology can help manage warranty information and claims processes more effectively, which can positively impact revenue. What is Warranty Management Software? Warranty Management software allows manufacturers to optimize their warranty management process by helping them create, monitor, process, and track warranties. Warranty management software enables users to stay on top of claims, coverage, and customer requests across the entire service life-cycle. Today’s warranty software products are also AI-enabled, allowing manufacturers to use machine learning capabilities to automate time-consuming functions. This helps improve the warranty process and ensures a superior level of customer service while delivering profitability. Benefits of a Warranty Management Software In general, warranty management software helps ensure uniform standards in warranty processes while providing increased transparency and communication between manufacturers, suppliers manufacturers. Let’s look at some of the additional specific benefits: Reduce warranty costs Technology can help manufacturers streamline processes in a way that can save time and shorten the service life-cycle, and as a result, reduce costs. APCQ (American Productivity & Quality Center) found that the more digitally mature an organization’s supply chain, the lower the warranty costs. Costs are found to drop from 3.5 % of sales when decisions are made based on analyzing past actions. Improve aftermarket excellence Reduced errors, faster resolutions, and accuracy in troubleshooting are the benefits of a warranty software solution, which improves customer satisfaction and builds a reputation for aftermarket excellence. Gain real-time intelligence Due to the digitization of many aspects of the manufacturing process, proper warranty management can offer manufacturers the ability to view in almost real-time the service experience and product usage across the entire service life-cycle. As data starts getting captured by more connected devices such as IoTs and automated warranty systems, manufacturers can analyze and avail near real-time insights into product performance and service status. Reduce fraudulent activity AI-based systems are using image recognition to identify fraudulent claims in a faster and more cost-efficient way. Machine learning algorithms are trained using thousands of images and can detect real issues against digitally manipulated images or past claims. Improve processes Warranty management software offers a closed-loop approach that can help optimize the claims procedure. By reducing operational discrepancies, warranty management can help manufacturers improve the process incrementally. Increase visibility between teams Often, teams across the service life-cycle don’t have visibility into warranty information, products, assets, and customer information. Access to service data can help all stakeholders have clear communication and visibility, improving collaboration. Optimize revenue Warranty management systems can help reduce losses incurred due to logistic delays or fraudulent claims. Time spent analyzing claims can also be minimized, saving resources and enhancing productivity, which can make warranty management a more cost-effective process. Tavant offers manufacturers an AI-driven, next-generation warranty management solution, which has helped organizations reduce warranty costs, increase supplier recovery, and improve aftermarket excellence. Talk to us for more information. SOURCES: https://tavant.com/products/warranty-management/ https://www.sdcexec.com/sourcing-procurement/article/21196030/apqc-metric-of-the-month-reducing-warranty-costs-as-a-percentage-of-sales https://www.idc.com/
Data Analytics: A Catalyst for Change in Service Life-cycle Management
The past few years have seen manufacturers look at their aftermarket services management in a completely new way. While technology and digitization have largely driven this change, the recent global pandemic has rocketed the drive for remote yet effective service support to ensure that customer requirement are still seamlessly met. Tech Innovations and the Flood Called Data The inadvertent result of this upsurge in digitization has been the data. Data, which is often collected from disparate sources, is now becoming a big challenge and an opportunity for manufacturers. With the adaptation of technology, many manufacturers can capture and utilize data but fail to do so. Why Measurement Matters Data-driven manufacturing is in the realm of being seen as a strategic necessity that can help manufacturers compete effectively. And with the application of analytics, manufacturers, suppliers, and distributors can achieve significant value in speed and operational efficiency. The ability to measure and use data is also leading manufacturers to offer services based on usage, uptime/downtime, and create value for customers through personalization. Let’s look at some of the key uses of data analytics and how it will impact manufacturers. Manage Demand and Supply Chains Data analytics is helping manufacturers understand the cost and efficiency of every aspect of the product lifecycle, from suppliers to customer usage. By analyzing the parameters and conditions that impact the supply chain from all angles, businesses can uncover problems such as hidden bottlenecks or unprofitable production lines. As a result, they gain insight into the conditions that affect the complete profitability of an integrated supply chain and learn how best to capitalize on given conditions. Forecast Demand for Products & Services Manufacturers can combine data with predictive analytical tools to create an accurate projection of purchasing trends. Insights driven by analytics can even help manufacturers understand how well lines are operating, enabling smarter risk management decisions. The ability to analyze when warranties are expiring can also result in additional service revenue channels for manufacturers. IoT solutions for asset management offer real-time alerts, enabling manufacturers to act quickly, and minimize losses from delayed, damaged, or lost goods. Proactive System Maintenance Predictive maintenance is helping manufacturers increase their product lifetimes while preventing downtimes. It analyzes the historical performance data to forecast potential failure and further identify the cause of the problem. This is particularly effective in field service management, where predictive maintenance can result in tremendous savings. According to McKinsey, manufacturers using predictive maintenance typically reduce machine downtime by 30 to 50 percent and increase machine life by 20 to 40 percent. Optimize Machine Efficiencies and Utilization Data analytics can significantly improve assembly-line efficiency by identifying bottlenecks and defects. With advanced analytics, manufacturers can ensure that machines operate at high efficiency, resulting in improved quality and increased productivity. Optimize Inventory and Warehouse Costs Efficiently Advanced analytics can be applied to improve product flow management, which positively impacts inventory operations while reducing unnecessary expenditure. For example, manufacturers can assess fill rates which can reduce stock-outs. Improved insights can help manufacturers know which locations/equipment are operating at an optimized level and improve other production centers and address warehousing deficiencies if any. Final Thoughts Enhancements Across the Service Life-cycle Analytics is enabling manufacturers to scale cloud-based operational intelligence, AI-enabled monitoring, diagnostics, and asset lifecycle management. AI-enabled digital technologies are seamlessly addressing service life-cycle challenges, increasing transparency across the process and functions, and creating a seamless and rich experience for the customers. SOURCES: http://www.wonderware.es/wp-content/uploads/2017/02/WhitePaper_InvensysandMicrosoft.pdf https://www.mckinsey.com/business-functions/operations/our-insights/manufacturing-analytics-unleashes-productivity-and-profitability
Futuristic Tech: Turning the Wheels of Manufacturing
As 2021 rapidly progresses, the impact of the COVID pandemic on all industries is being felt. Businesses have had to maintain employee/customer distances resulting in the growth of communication-enabled technologies. The manufacturing industry has also been similarly impacted. While technology has always been an enabler, today, we are beginning to see advancements in ways that interconnect humans with smarter machines for improved processes, performance, and protection. WEARABLES Wearables are being used in manufacturing to ensure improved safety conditions in work environments. Wearables such as wristbands, smart clothing, or headwear like Google Glasses can provide information to the wearers such as surfaces that are hot, short-circuited equipment, machine malfunction, or even hazardous spills. Wearable technology for construction workers may also soon help reduce construction worker fatalities and injuries. A U.S. Chamber of Commerce report notes 23% of contractors will be using such technology by 2021. Wearables are also helping managers gather data that can help towards improving efficiencies and optimization. For example, it may be observed that time is spent by workers using a forklift for short distances between two essential work points, which can be reduced if the floor plan is reworked or eliminated by a conveyor belt. IOTs The manufacturing industry has already deployed IoTs extensively to collect essential production data and turn it into valuable insights. IoTs have helped manufacturers improve operational efficiency and reduce delivery times. The PwC’s 2019 Internet of Things Survey reveals that manufacturers are optimistic about the benefits of IoTs with 68% planning to increase their investment over the next two years. By deploying enhanced IoTs that are sensor-enabled (audio, video, temperatures, vibration, voltage), manufacturers are now also staying ahead of potential machinery issues and can perform predictive maintenance for improved customer satisfaction. Recently a Boston-based contractor used an algorithm that analyzed photos from job sites, scanning them for safety hazards and correlated with past accident records. Construction companies can therefore potentially compute project risk and know which projects have higher threats. DRONES While there are still technological, organizational, and regulatory challenges to implementing drones, manufacturers have begun experimenting with drones in warehouse operations and inspection tasks. Drones are being used to monitor and connect the stages of the manufacturing process, such as moving components to a production line, inspection, or delivery of the final product to shipping. And when regulatory limits are removed, many manufacturers are looking to deploy drones to help them with field inspection and logistical tasks. AR/VR TECHNOLOGY Virtual reality can help manufacturers test and enhance products digitally without creating expensive prototypes, saving time and money. Automobile manufacturers are already using virtual reality to ensure cars are tested at an initial phase of the vehicle development process, reducing the time and cost involved in ensuring tolerances and safety and altering the design features. In construction, custom workstations are being built using virtual reality to offer teams immersive design review and collaboration capabilities. Augmented reality can be used to monitor field conditions, measure changes, and help manufacturers envision a finished product. SMART ROBOTICS Industrial robots have been speeding up manufacturing operations for the past decade. In fact, in another recent PwC report, 59% of manufacturers are already using some form of robotics technology. Today, however, we are beginning to see more AI-enabled robots that collaborate with human workers. The Tesla Gigafactory uses smart self-navigating, Autonomous Indoor Vehicles (AIVs) to shift goods between workstations. Companies like Cornell Dubilier, a power capacitor manufacturer in the US, also use ML-trained robots to inspect capacitor installations, doubling its speed of labelling process from 125 parts an hour to 250 parts an hour. AI AND MACHINE LEARNING AI and machine learning make it possible for manufacturers to improve processes and products through intelligent feedback, which the algorithm can constantly learn from. According to Deloitte, machine learning improves product quality up to 35% in discrete manufacturing industries. Even the construction industry (considered one of the most under-digitized industries in the world) uses AI to predict cost overruns based on project size, contract type, and the competence levels of project managers. AI-driven cameras also help construction workers avoid spending hours walking around searching for tools, as AI can immediately recognize and locate on-site tools and equipment. Advanced machine learning systems offer smarter decision-making capabilities to manufacturers and can improve tasks such as research, development, and product line extension. 3D PRINTING While advances in 3D printing have helped streamline prototyping, one of the most encouraging outcomes is its potential for mass customization. The 3D printing industry is projected to reach USD 63.46 billion by 2025 and is offering manufacturers the ability to innovate through its introduction of new materials. Unbelievably, 3D printing also offers builders the ability to produce entire houses! Start-up 3D printing construction company, Icon says that 3D printing can reduce construction costs by up to 30% and produce a home twice as fast as traditional methods. BIG DATA AND ANALYTICS The adoption of sensors and connected devices have resulted in a tremendous increase in the data points being generated for the manufacturing industry. Only by applying advanced big data analytics, can manufacturers use this vast information to discover insights and identify patterns. When properly deployed, analytics can help manufacturers improve processes and supply chain efficiency and predict the variables that could adversely affect production. Additionally, big data and analytics can be applied to assess damages in buildings and fraud detection. Fraud analytics using machine learning to prevent contractors from making false claims by analyzing images against existing claim databases. This is particularly useful in construction such as roofing, which needs to be viewed from above and can easily be faked. Advanced analytics can automate the processing of roof condition and minimize the need for aerial imagery which be expensive, time-consuming, and unsafe. THE FUTURE IS IMPROVEMENT, NOT DISRUPTION Successful manufacturers are often the businesses that can orchestrate and align all the facets of their operations smoothly. Therefore, the use of technology in manufacturing is mainly driven by their need for better efficiency and control over
Bayesian Methods for Media Mix Modeling with Carryover and Shape Effects
As discussed in my previous blog posts, a lot of research is being done in ad attribution and media mix modeling. Today I’ll introduce another paper that provides some interesting analysis. Fair warning, you should have a basic idea of Bayesian regression before reading this. You can find a great introduction here. Carryover and Shape Effects The authors’ most exciting contribution is incorporating carryover and shape effects in their media mix model. Carryover effects try to model the impact of media spend over a future period. Since media spend influences consumers on buying a product or service, the impact of such spending doesn’t just last for the time an advertisement is aired but for a more extended period. The authors transform the time series of media spend using a decay function for accounting for such carryover effects. They use the adstock function as described below : wm is a non-negative weight function, and the media spend effect is the weighted average of media spend of the current period and previous L-1 periods. The authors introduce two types of weight functions, geometric decay (where media spend peaks when an advertisement is aired) and delayed adstock (where the impact of media spend rises sometimes after an ad is aired). A visualization describing the effects of the weight functions can be seen below, taken from the paper. Next, the authors discuss shape effects. Shape effects aim to capture diminishing returns on media spend. For example, it is valid to assume that for a specific medium, the rate at which media spends rises is dramatic from 0 to $50 but reduces significantly from $100 to $150. The authors use a Hill function to model shape effects. The discussion of Hill functions is beyond the scope of this blog, but the regression coefficients can be multiplied by the Hill function to get the following form : The hill function for media spend is a point transformation, as opposed to earlier discussed carryover effects. The following graph, taken from the paper, gives a visual representation of diminishing returns, given different parameter values in the Hill function : Both these transformations can be applied to media spend. Depending on individual use cases, one must decide which transformation to apply first. The authors apply the adstock transformation first and then the shape transformation. The final sales at time t, which can be described as y_t , can be modeled using the following equation : To simplify, this equation models sales as a function of some baseline sales τ in addition to transformed media spend effects of control variables, and random noise. Why Bayesian Regression A common question could be: Why estimate these parameters using bayesian regression? The answer lies in the fact that Bayesian regression lets us quantify the uncertainty in our predictions, and more importantly, allows us to set priors on our parameters. For example, it is valid to assume that media spend will never have a negative effect on sales, which allows us to set informative priors on media spend coefficients (constraining them to be non-negative values). The authors then explain their implementation of this model to real-world datasets. They use Gibbs sampling to sample from their model and implement this in STAN. However, multiple techniques for sampling from the posterior distribution and their code can be replicated easily using PyMC3. Please take a look at the fundamentals of Bayesian Regression if this isn’t making much sense. The parameter estimates obtained from the model can then be plugged into a linear optimization algorithm that conditions on a fixed media spend budget to find the best media mix given a set of channels. The linear optimization algorithm introduced by the author is beyond the scope of this post, but I might discuss it in my next one. Stay tuned!
How Cloud Technology Can Leapfrog Your Business IQ
Agility. Innovation. Intelligence. When working in tandem, these three concepts can spell success for any business. But without these capabilities, businesses falter in the eddies of unfavorable market conditions. When business leaders get together, the discussion often gravitates to how companies can nurture and empower new ideas and implement them faster because this is what keeps any business competitive. Business Success…and the stumbling blocks called data Across the globe, enterprises are investing heavily in AI, ML, and technologies for improved efficiency, in solutions that communicate with each other seamlessly, and in strategies that empower employees to ideate and innovate. And yet, there’s still a hurdle that most companies continue to stumble over: data. A recent IDC research into enterprises showed that 50% of employees are overwhelmed by the amount of data, while at the same time, 44% say they don’t have enough data to support decision making! What cloud technology can mean to business growth Businesses everywhere are going through a period of transformation. Clunky old legacy systems are being digitally overwritten. Decades-old data silos are shrinking under a barrage of data management solutions. And all of these transformative capabilities are being held up and supported by the cloud. More than a storage system If you’ve been thinking of the cloud as an efficient way to store data, it’s time to upgrade that thinking. The cloud is a platform that can handle data and while also supporting the solutions that process and use that data. By leveraging cloud capabilities, businesses can scale their capabilities up or down with less risk and pursue their business goals while keeping costs low. No downtime Imagine a credit reporting company trying to move 18 years of customer data for 330 m American customers US from GVAP archives. By using cloud technology Tavant was able to dynamically adjust the size of the computing cluster to accommodate the changing workload without interrupting production. Remote access and security The global pandemic has put greater focus on cloud capabilities as companies are forced to maintain data security while enabling employees to work remotely. Consider the competitive advantage experienced by businesses that already had a secure, remote work environment in place before 2020. CAAS (containers as a service) Businesses can also benefit tremendously from container applications which are now being offered by many cloud providers. As consumable services, these CAAS can be deployed by DevOps directly on top of the cloud application layer. Each app is wrapped in a standardized configuration, significantly improving security, scalability, and load times and providing an efficient alternative to virtual machines. In fact, Gartner predicts that by 2023, 70% of global organizations will be running more than two containerized applications in production, up from just 20% in 2019. Smart businesses are getting smarter with cloud technology Regardless of the business size, cloud technology now offers an easy way for businesses to innovate, respond quickly and empower employees from anywhere. For the first time ever, we have a level playing field from which companies of any size can leapfrog their way into the future and create new business opportunities for themselves using cloud technology. The only question is, who will leverage cloud capabilities efficiently to get there first? SOURCE: https://blogs.idc.com/2020/05/15/defining-the-data-native-worker-gen-d/ https://www.entrepreneur.com/article/345826
Building Trustworthy and Ethical AI is everyone’s responsibility
Whether you realized or not, Artificial Intelligence (AI) has quickly become part of our daily life. With traditional industry and businesses like fintech, media, healthcare, pharmaceuticals, and manufacturing adopting AI rapidly in recent years, concerns related to Ethics and Trustworthiness have been mounting. Today, AI ‘assists’ many critical decisions influencing people’s life and well-being, for example, creditworthiness, mortgage approval, disease diagnosis, employment fitment, and so on. It was observed that even with human oversight, complex AI systems may end up doing more societal harm than social good. Building Trustworthy and Ethical AI is a collective responsibility. We must apply fundamentals throughout the lifecycle of AI, for example, product definition, data collection, preprocessing, model tuning, post-processing, production deployment, and decommissioning phases. No doubt Government and Regulators have a role to play through monitoring and ensuring a level playing field for everyone, the same is for people building, deploying, and using AI systems. This includes executive leadership, product managers, developers, MLOps engineers, data scientists, test engineers, HR/Training teams, and users. Bias and unfairness While Trustworthy and Ethical AI is a broader topic, it’s tightly coupled with the prevention of of Bias and Unfairness. As the National Security Commission on Artificial Intelligence (NSCAI) observed in a recent report: “Left unchecked, seemingly neutral artificial intelligence (AI) tools can and will perpetuate inequalities and, in effect, automate discrimination.” AI learns from observations made on past data. It learns the features of data and simplifies data representations for the purpose of finding patterns. During this process, data gets mapped to lower-dimensional (or latent) space in which data points that are “similar” are closer together on the graph. To give an example, even if we drop an undesired feature like ‘race’ from the training data, the algorithm will still learn indirectly through latent features like zip code. This means, just dropping ‘race’ will not be enough to prevent the AI learning biases from the data. This also brings out the fact that data ‘bias’ and ‘unfairness’ reflect the truth of the society we live in. With not enough data points belonging to underrepresented sections of the society, high chances that they will be negatively impacted by AI decision-making. Moreover, AI will create more data with its ‘skewed’ learning which will be used to train it further and eventually create further disparity through its decision-making. Trustworthy and Ethical AI is important By definition, Trustworthiness means “the ability to be relied on as honest or truthful”. Organizations must ensure their AI systems are trustworthy, in absence of trust, undesired consequences may occur, including but not limited to business, reputation and goodwill loss, lawsuits, and class actions that can be potentially life-threatening for a business. On the other hand, Governments and Society must ensure that AI systems follow Ethical principles for the greater good of common citizens, one great example is UNESCO Ethical AI Recommendations. As per the European Commission Ethics Guidelines for Trustworthy AI, Trustworthy AI must be Lawful, Ethical, and Robust. Respect for human autonomy, fairness, explicability, and prevention of harm are four critical founding principles of Trustworthy AI. It’s critical that AI should work for human wellbeing, ensure safety, should be always under humans’ control, and never ever should harm any human being.. Who is driving Ethical AI? Realization of Trustworthy AI is envisioned through the following actions: Who is driving Ethical AI? Leading tech companies have already announced one or another kind of Ethical AI initiatives and governance. As there is no common ground in terms of benchmark principals, guidelines, and framework, it’s difficult to assess whether the intent is genuine or merely optics. As AI will have a profound impact on society and well being of common citizens, just ‘self-certification’ will not be enough. Governments should (some have started already) define the principles, policy, guidelines and establish an effective oversight and regulatory mechanism. This will help to ensure that common citizens are protected from intended/ unintended negative fallouts of AI. As AI evolves, frameworks and regulations should also evolve. Recently, the US Federal government signed Executive Order On Advancing Racial Equity and Support for Underserved Communities, however, more needs to be done. EU, UN & DoD have already taken the lead on this topic, with European Commission Ethics Guidelines for Trustworthy AI, UNESCO Elaboration of a Recommendation on the ethics of artificial intelligence and US Department of Defense Ethical Principles for Artificial Intelligence should be considered as baseline work towards defining a practical and mature guideline towards Trustworthy and Ethical AI. Plan of action Here we attempt to identify suggested actions for involved actors. This is in no way an all-inclusive list and should be taken as only a baseline and should be updated to support a particular case: Conclusion We all have actions to build Trustworthy and Ethical AI for the larger good of society (and humanity). With coordinated and persistent efforts, it is definitely possible.
Causally Motivated Attribution for Online Advertising
As mentioned in the previous blog post, algorithm-based methodologies for assigning credit to media channels on conversion of a user are becoming more and more popular, replacing archaic methodologies such as first touch and last touch attribution. A paper that goes beyond a regression framework to explain such attributions was presented by Dalessandro et al. which I’ll be going over in the next few sections. Attribution Attribution and Causality Dalessandro et al. propose a counterfactual analysis to produce estimates of the causal effect of advertising channels on user conversion. There are some strict assumptions that have to be met in order to obtain causality from the data, which Dalessandro et al. state as the following: The ad treatment precedes the outcome (conversion of a user) Any attribute that may affect both ad treatment and conversion outcome is observed and accounted for. i.e., there are no unknown variables acting as confounders. Every user has a non-zero probability of receiving an ad treatment. Obviously, in real life scenarios, conditions 2 and 3 are nearly impossible to prove as true in any attribution analysis. It may be possible that an ad campaign is targeted towards a certain demographic, thus violating condition 3, and it may be very possible that confounders such as users’ biases towards certain products and services are unmeasurable quantities. One can see how this would be a challenge. In the interests of brevity, we will not dwell on the mathematical formulation of such an analysis since the practicality of it is dubious. In the next section, I will discuss an approximate causal model that Dalessandro et al. introduce, which recasts the causal estimation problem as a channel importance problem, with better application to real world data. Channel Importance Attribution Before getting into any convoluted equations, I’ll quickly introduce important notation: C={ C1, C2,…Ck } is defined as the set of media channels that have shown ads to a group of people W is a vector of user attributes before being exposed to any ads ( for example, demographics, prior internet searches etc.) Y is a boolean indicating whether or not a user has converted, post exposure to ads (γ = Σ Y, n) is the dataset of n users who have seen the same ads by channels in C, and have the same values W = w, producing γ = Σ Y total conversions S is the set C, excluding Ck (hence a subset of C) ωS,k is the probability that set C begins with the sequence {S, Ck, ….} in some distribution Ω of possible orderings The expectation of channel Ck‘s contribution to Y, over all possible combinations of C, is given as Vk, which can be seen in the equation below: In order to understand this better, consider an example where there are only 2 channels, C1 and C2. Attribution values for the channels can be given as : We can see in this simplified form that the attribution values are affected by how these channels serve their advertisements to the user. It is interesting to note, that in the case of observable ad campaigns, we will already know the order in which channels deliver their ads, making the ωS,k probabilities always 0 or 1. The paper discusses why this observable information can actually be harmful in providing attribution values. Let’s take a look at an example. Consider C = {C1, C2}. Further, let E[γ|{∅}] = E[γ|{C1}] = E[γ|{C2}] = 0, and E[γ|{C1,C2}] = δ >0.. Further, assume that C2 always serves its ads after C1. These assumptions tell us that the individual effects of C1 and C2 cause no conversions among users, but the joint effects of C1 and C2 do lead to some user conversions. Using the formula described above, we can get attribution values as following: Since we have observable probabilities of the sequence in which the channels serve their ads (since C2 always serves after C1), we can note that ω2,1 = 0, and ω1,2=1, giving us the equation in the form above. What is interesting to note now, is the fact that our attribution values tell us that V1 = 0, while V2 = δ. This means, all the credit for the joint effect of C1 and C2 in our example is going to C2, simply due to the fact that C2 serves its ads after C1. This conclusion is harmful, since we can extrapolate this to a general idea that channels that serve their ads later receive greater credit for user conversions ( it basically turns into a last touch attribution model, which is pretty flawed). Dalessandro et al. recognize that using these observable probabilities lead to poor recognition of interaction effects among channels, and instead propose a different way to calculate the quantity ωS,k. The following equation is the crux of their idea : They define Ω as a uniform distribution over all possible orderings of C. They state that ωS,k can now be calculated as : To completely understand this equation would require a very good understanding of Shapley Values, which are a common concept of attribution allocation in game theory. Due to the limited scope of this blog, I will not discuss it here. But if there’s something to take away from the paper’s implementation, it is the fact that observable probability distributions of ωS,k should be ignored in favor of the equation provided by the authors in the equation above.
The forensic goldmine of smart television viewing
In the United States, the total smart TV household penetration has been increasing rapidly from 2012 onwards, from around about 9% in 2012 to 60% in January 2020. This growth comes from shifting consumer preference towards online content. The wide availability of high-speed internet and the smart features of connected TVs have also contributed to the fast growth of the smart TV market both in the United States and the world. And, added to that, the pandemic has further pushed the consumption of content up even further. Americans spend an average of 3½ hours in front of a TV each day, according to eMarketer, the market research company. With more and more consumers opting for Smart TVs, marketers and publishers today have access to a wealth of consumer information.
The Forensic Goldmine of Smart Television Viewing
In the United States, the total smart TV household penetration has been increasing rapidly from 2012 onwards, from around about 9% in 2012 to 60% in January 2020. This growth comes from shifting consumer preference towards online content. The wide availability of high-speed internet and the smart features of connected TVs have also contributed to the fast growth of the smart TV market both in the United States and the world. And, added to that, the pandemic has further pushed the consumption of content up even further. Americans spend an average of 3½ hours in front of a TV each day, according to eMarketer, the market research company. With more and more consumers opting for Smart TVs, marketers and publishers today have access to a wealth of consumer information. Content Recognition & Targeting Automatic content recognition (ACR) technology has the potential to capture all types of TV viewing: linear, OTT, video on demand, commercials, and video games. When tracking is active, Smart TVs can record and send out everything that comes up on the screen regardless of whether the source is cable, an app, the DVD player, or a set-top box, but without personally identifiable information. Once collected, media analytics companies consume the ACR data, then clean, compare and combine it with other data sets to make it more usable and accurate. TV advertisers, therefore, no longer need to rely on Gross Rating Points (GRPs) and have greater capabilities of showing their ads to the right person at the right time. Analytics & Advertising While the world of advertising is moving to digital, TV advertising still accounted for $84 billion in 2018 in the US alone. But data-driven methods are enabling these dollars to be spent more efficiently using automated systems over programmatic TV. Advanced advertising technology enables advertisers to have more control with end-to-end inventory visibility, audience, and demand. Using specially developed software solutions, marketers can integrate and streamline omnichannel advertising and marketing activities. Advanced analytics technology for advertising can help businesses also use ACR data to connect ad spend to business goals, like driving in-store traffic and make intelligent media advertising plans. Data-Driven Media Subscription Management Subscription rates after March 2020 grew between 3 times for digital news and up to 7 times for streaming services as published in the Covid19 Subscription Impact Report conducted by Zuora. By analyzing television viewership data in conjunction with product subscription information, publishers can manage subscription features specific to OTT platforms, such as auto-renewing and churn management. Netflix has claimed that its media recommendation solution could be saving up to $1 Billion a year by decreasing churn. AI-based Viewership Recommendations Content metadata in smart televisions are often only applicable to an on-demand video where there is time to generate it before distribution. Advance recommendations based on prior knowledge are irrelevant in cases like live sports, where viewership is based on expectations instead of prior information. For this reason, operators need to leverage AI/ ML to generate effective recommendations, even for VoD content. Recommendation engines help uncover video content for users that they would not be likely to find themselves. As a result, video and TV services can increase their content reach without having to constantly acquire new content. Tavant specializes in advanced advertising technology and media analytics to help companies gain the most from smart tv data. For more information on how we can help you write to us at [email protected]; or click here. SOURCES: https://www.washingtonpost.com/technology/2019/09/18/you-watch-tv-your-tv-watches-back/ https://dl.acm.org/doi/10.1145/2843948