This article was originally published in Analytics India Magazine. What forms the core of businesses today? Huge volumes of data that flow in and out every day — and though it does matter, what really comes into play is the ability to use data and models to make better business decisions. Sai AlluriAnalytics Lead at Uber India talks about supply positioning models, segmentation and visualization tools that are applied at Uber, and how Uber stays on top of the game by understanding the biggest mismatch between supply and demand.
The supply positioning model at Uber refers to anticipating demand patterns, and placing driver partners across those hubs with the aim to plug in the demand, lower ETAs and increase overall efficiency. One of the key focus areas is moving from a passive supply-positioning model to act through specific recommendations across the network. In the words of Alluri — supply optimization is one of the biggest focuses at Uber and the challenge is to efficiently optimize the supply wherever there are high areas of demand or can be.
Say for example, when you see a search surge multiple in 2x or 3x, it portrays how much demand is in that particular area and what supply would you need to meet this demand.
It is very difficult for a driver to move from Gurgaon to CP given the traffic conditions and it might take him longer to reach. How do we know in advance where this demand is going to be based on historical data? Uber does supply positioning by specifically a breaking down the city into multiple pockets, b then identifying these pockets based on the demand parameters that show up, c once you identify these pockets, you can figure out how you want to position the supply chain in these specific areas.
Key parameters addressed for the analysis are: broken up by the hour of the day, by day of the week and by the specific pocket.
So now that you have the information, how do you use it to inform future decisions? In case of Uber, the real challenge is in filling the demand-supply gap. The motive is to set a benchmark and rule out weekly anomalies.
And it is further used to build a potential forecasting model, where one can predict the highest demand or lowest supply and keep modifying it on a weekly or bi-weekly basis as the data changes.
At Uber, the goal is to drive efficiency across all areas of business. For example, as soon as the driver becomes active on our system, we want to make sure if he has any questions pertaining to how do you go online or how do you essentially go pick up your customer they are answered.
Understanding Supply & Demand in Ride-hailing Through the Lens of Data
The communication dispatch was targeted at converting drivers into loyal Uber partners. Clustering Analysis basically means breaking up huge data sets into further subsets to help get better insights into critical decision areas.Dynamic Pricing - Uber Prices
In this case, segmentation was based on hours and trips. Alluri shares how the model was further optimized to include trips and how it led to increased revenue for drivers. Visual analytics is used at Uber to make data look more actionable and understandable.
The team uses visualization layers for most business insight applications and uses it to find out the sequence of data flowing in.1988 camaro tachometer
He worked in consulting before joining Uber in San Francisco, California. Sai shifted to India last year to set up a team and focus on operational and analytical challenges in India. He is part of the industry professionals team working closely with UpGrad to create a world-class learning experience.
Want to know more about this case-study or various other real-life examples from many other industry leaders who have partnered with UpGrad? Or hear from one of our students and his experience on how the UpGrad Program changed his life.You are missing out if you have not tried Uber or Lyft. They are the easiest, most convenient, and least expensive way to get a ride.
In case you missed it, check out this article on Uber that we wrote last year. I will refer only to Uber throughout the rest of this article because I am most familiar with their services, but Lyft has its share of fans and has recently gained marketshare from Uber.
I started using the service fairly regularly, and I always ask the drivers whether they like their jobs and are making decent money. Another email followed a few months later with the same message.
Uber drivers are independent contractors who drive their own cars, pay their own expenses gas, maintenance, insuranceand do not receive any guaranteed income or benefits. I was always a bit skeptical when Uber drivers would tell me how much they made.
I figured that they must be underestimating their expenses. The average earnings per driver varied considerably by city. The drivers also earn some tips, which are not included in the above figures. Tipping is not encouraged by the company, but drivers sure appreciate them. Check out our article on the history of tipping.
All earnings are after expenses, which were estimated by the company based on the following assumptions which seem low to me :. I always figured that I could become a billionaire if I could just figure out a business that harnessed the talents and work ethic of all of the stay-at-home moms, retirees who want to work part-time, and underemployed who want to moonlight to make ends meet.
Such a company would have to provide a valuable service and enable team members to work flexible hours. Sadly, I never came up with such an idea, but Uber did. It came up with the perfect business model, and the rest is history. The company is now worth more than General Motors.
The flexibility of being an Uber driver led to a seemingly endless supply of people who wanted to drive for them. In a free market economy, when supply goes up, prices fall.The world needs independent businesses. Uber is an on demand transportation service which has brought a revolution in the taxi industry all across the world. The business model has made it possible for people to simply tap their smartphone and have a cab arrive at their location in the minimum possible time, leaving a lot of budding startups yearning for an App like Uber.
Read on! Founders : Travis Kalanick and Garrett Camp. Number of Users : More than 50 Million. Number of registered drivers : Approximately 7 Million as of November Average number of daily Uber Trips : 1 Million.
If you have ever travelled in a taxi, you might have paid the driver in cash at the end of your journey. The cash collected by each journey is the only source of revenue for a traditional cab company.
Uber is no different. Neither does Uber have a different revenue model than the one mentioned above nor it has any other source of revenue as of now.
But just imagine 1 million rides a day. What has made them so successful is the fact that the revenue model is as unique as their business model.
It can be explained as:. Different cab models to cater to everyone :. Uber has not limited itself to a particular segment of cars or to a particular segment of people. Surge Pricing Technology :. Variation in cab fares according to situation is an important aspect of their business model.
Whenever the demand increases, per mile prices are automatically increased. The new price depends on the number of available drivers and the number of requests made by people who want to travel.
It has applied for a price surge technology patent in the US. Other Uber rides :. Uber has come a long way from cabs. It now offers boats, helicopters as well as some other transportation means on demand. Uber has such a vast customer segment that it has got something on offer for everyone. Uber serves professionals as they hire an Uber cab to and fro work. For this Uber did few tie-ups with corporates in the beginning and does so when it launches in a new city in a new country.
Apart from professionals, Uber tries to touch hearts of people by offering special services like:. Uber for Kids : A special service from Uber dedicated for parents who want their kids to reach home from school in an Uber cab. Uber for Senior Citizens : Another special service from Uber where it targets senior citizens. This made Uber have some special features for seniors and hence attract more senior citizens on the platform.
Check out this post from fortune: Why senior citizens are flocking to Uber. All this might raise another question in your mind. The question about how does Uber find them or how does it market out to its target audience?
We extended our research beyond Uber business model and came up with an entire growth model of Uber and some insights about how you can build an Uber like App.
How Uber Uses Data Analytics For Supply Positioning & Segmentation
In less than 6 years, Uber has managed to become the best example of a city-by-city mobile service company roll-out. The underlying principle here is that for every city it launches, it faces the same chicken and egg problem.By unifying the teams in her organization — the data science ninjas, as she calls them — with their cross-functional counterparts in engineering, product, and design, Uber has built internal tools and platforms that others in the company can leverage, bringing cutting-edge expertise to anyone within the company.
How far do you want to forecast? Everything else is done completely underneath the hood. Before developing the platform, Bell mapped out the three questions the team examines to determine if the investment in any kind platformization will be worthwhile.
Number one: Is this area of data science going to greatly enhance the user experiences? Two: Are there multiple different use cases across the company? Three: How reusable are the various different methodologies or modules from use case to use case?Outlook stopped working close the program
Forecasting was a slam dunk across the board, and she went on to explain its many use cases across Uber, including supply and demand, real-time outage detection, and hardware capacity planning. With the vast number of metrics, paired with the fact that many of these are supply- and demand-driven, it would be impossible to set up those thresholds by humans and keep them up-to-date over time.
The company gets hundreds of millions of signals continuously, whether from back end systems it is tracking or marketplace health indicators. Hardware capacity planning formed the third leg of forecasting, using the platform to monitor closely how much hardware to purchase and provision for.
These are very much day-of-the-week dependent. Leveraging its forecasting expertise, Uber has now built a completely automated tool said Bell. Check out all the other sessions you missed at Transform as well. Lockdown got you stressed? Now's the perfect time to try meditation with this top-rated app.
Stay busy while social distancing with a lifetime pass to Rosetta Stone and more for a huge discount. Stuck inside? View all deals.What forms the core of businesses today?
Huge volumes of data that flows in and out every day — and though it does matter, what comes into play is the ability to use data and models to make better business decisions. UpGrada leading edtech startup, have collaborated closely with Uber for their Data Analytics course content generation.
Sai Alluri, Analytics Lead at Uber Indiaand Industry Expert at UpGrad, talks about how supply positioning model, segmentation and visualization tools that are applied at Uber and how Uber stays on top of the game plan by understanding the biggest mismatch between supply and demand. The supply positioning model at Uber refers to anticipating demand patterns, and placing driver partners across those hubs with the aim to plug in the demand, lower ETAs and increase overall efficiency.
One of the key focus areas is moving from a passive supply-positioning model to active through specific recommendations across the network. In the words of Alluri — Supply optimization is one of the biggest focuses at Uber and the challenge is to efficiently manage optimizing the supply wherever there are high areas of demand can be.
One of the methodologies is through searchsurge — in real time, meaning that supply comes in from the highest area of demand. Say for example, when you see a search surge multiple in 2x or 3x, it portrays how much demand is in that particular area and what supply do you need to meet this demand.
Uber analyzes historical data for say, last three or four weeks and identifies pockets within the city that witness extremely high demand. It is very difficult for a driver to move from Gurgaon to CP given the traffic conditions and it might take him longer to reach. How do we know in advance where this demand is going to be based on historical data? How Uber does supply positioning is by specifically a breaking down the city into multiple pockets, b then identifying these pockets based on the demand parameters that show up, c once you identify these pockets, you can figure out how you want to position the supply chain in these specific areas.
Key parameters addressed for the analysis are: broken up by hour of day, by day of week and by specific pocket. So now that you have the information, how do you use it to inform future decisions? In case of Uber, the real challenge is in filling the demand supply gap.
The motive is to set a benchmark and rule out weekly anomalies. And it is further used to build a potential forecasting model where one can predict the highest demand or lowest supply and keep modifying it on a weekly or bi-weekly basis as the data changes. At Uber, the goal is to drive efficiency across all areas of business. For example, as soon as the driver becomes active on our system, we want to make sure if he has any questions pertaining to how do you go online or how do you essentially go pick up your customer.
The communication dispatch was targeted at converting drivers into loyal Uber partners. Clustering Analysis basically means breaking up huge data sets into further subsets to help get better insights into critical decision areas. In this case, segmentation was based on hours and trips. Alluri shares how the model was further optimized to include trips and how it led to increased revenue for drivers. Visual analytics is used at Uber to make data look more actionable and understandable.
Our team uses visualization layers on most business insight applications and uses it to find out the sequence of data flowing in. He worked in consulting before joining Uber in San Francisco, California.
He shifted to India last year to set up a team and focus on operational and analytical challenges in India. Bhasker is a Data Science evangelist and practitioner with proven record of thought leadership and incubating analytics practices for various organizations. With over 16 years of experience in the area of Business Analytics, he is well recognized as an expert within the industry. He is B. You must be logged in to post a comment. How Uber uses data analytics for supply positioning and segmentation.
How is supply positioning done at Uber?
Bhasker Gupta Bhasker is a Data Science evangelist and practitioner with proven record of thought leadership and incubating analytics practices for various organizations.
Share This.The tension between trying to standardize an experience and giving drivers true independence has manifested in two lawsuits.
Uber Technologiesa pending class-action lawsuit by Uber drivers in California, which alleges that drivers are misclassified as independent contractors. Uber argues that its business model is premised on licensing software that acts as an intermediary between passengers and drivers.
The second is a new lawsuit, Meyer v. Uber says that these claims are unwarranted. District Judge Jed Rakoff recently denied a motion by Kalanick to have the lawsuit dismissed. These cases clearly point to the importance of understanding how platform-employers operate, and how work is structured within semi-automated, algorithmic management systems.
Uber drivers interact with the platform. We analyzed online driver forums, where tens of thousands of drivers share advice and compares notes on their experiences and challenges with the Uber system. We also conducted in-depth interviews with seven drivers to explore worker experiences of the on-demand economy. The platform redistributes management functions to semiautomated and algorithmic systems, as well as to consumers.
Algorithmic management, however, can create a deal of ambiguity around what is expected of workers — and who is really in charge. Uber sets the rates. Uber has full power to unilaterally set and change the fares passengers pay, the rates that drivers are paid, and the commission Uber takes. Uber sets the performance targets. Customers act as managers. After every ride, passengers are prompted to rate drivers on a 1-to 5-star scale. This feedback generates instantaneous and recurrent performance evaluations that allow Uber to track worker performance and intervene with poor performers.
In order to remain active on the system, drivers must meet an average rating target that hovers around 4. Though rating systems can build and scale trust and accountability in platforms, they have their flaws. Discrimination may also be of concern, as consumers can directly assert their preferences and their biases in ways that companies are prohibited by law from doing.
To achieve good ratings, drivers must modify their behavior to produce a fairly homogenous Uber experience. The company encourages uniform behavior in a few ways. And the company routinely sends messages to drivers that explain how passengers rate particular behaviors. This redistribution of managerial oversight and power away from formalized management and toward a triadic relationship between employers-workers-consumers is part of a broader trend in the on-demand economy, and in the service industry more generally.
Uber suggests the schedule. It goes into effect when demand passengers outstrips supply drivers by a particular threshold.This article is the first in a series dedicated to explaining how Uber leverages forecasting to build better products and services. In recent years, machine learning, deep learning, and probabilistic programming have shown great promise in generating accurate forecasts. In addition to standard statistical algorithms, Uber builds forecasting solutions using these three techniques.
Below, we discuss the critical components of forecasting we use, popular methodologies, backtesting, and prediction intervals.
Forecasting is ubiquitous. In addition to strategic forecasts, such as those predicting revenue, production, and spending, organizations across industries need accurate short-term, tactical forecasts, such as the amount of goods to be ordered and number of employees needed, to keep pace with their growth.
Not surprisingly, Uber leverages forecasting for several use cases, including:. The Uber platform operates in the real, physical world, with its many actors of diverse behavior and interests, physical constraints, and unpredictability. Physical constraints, like geographic distance and road throughput move forecasting from the temporal to spatio-temporal domains. Figure 2, below, offers an example of Uber trips data in a city over 14 months. You can notice a lot of variability, but also a positive trend and weekly seasonality e.
You may notice that weekends tend to be more busy.Gauss seidel function python
Forecasting methodologies need to be able to model such complex patterns. Apart from qualitative methods, quantitative forecasting approaches can be grouped as follows: model-based or causal classical, statistical methods, and machine learning approaches.
Model-based forecasting is the strongest choice when the underlying mechanism, or physics, of the problem is known, and as such it is the right choice in many scientific and engineering situations at Uber. It is also the usual approach in econometricswith a broad range of models following different theories.
When the underlying mechanisms are not known or are too complicated, e. Popular classical methods that belong to this category include ARIMA autoregressive integrated moving averageexponential smoothing methods, such as Holt-Winters, and the Theta methodwhich is less widely used, but performs very well.
Recurrent neural networks RNNs have also been shown to be very useful if sufficient data, especially exogenous regressors, are available. Typically, these machine learning models are of a black-box type and are used when interpretability is not a requirement. Below, we offer a high level overview of popular classical and machine learning forecasting methods:.
Interestingly, one winning entry to the M4 Forecasting Competition was a hybrid model that included both hand-coded smoothing formulas inspired by a well known the Holt-Winters method and a stack of dilated long short-term memory units LSTMs.
Actually, classical and ML methods are not that different from each other, but distinguished by whether the models are more simple and interpretable or more complex and flexible.
In practice. At Uber, choosing the right forecasting method for a given use case is a function of many factors, including how much historical data is available, if exogenous variables e. The bottom line, however, is that we cannot know for sure which approach will result in the best performance and so it becomes necessary to compare model performance across multiple approaches.
It is important to carry out chronological testing since time series ordering matters. Experimenters cannot cut out a piece in the middle, and train on data before and after this portion.
Instead, they need to train on a set of data that is older than the test data. With this in mind, there are two major approaches, outlined in Figure 4, above: the sliding window approach and the expanding window approach. In the sliding window approach, one uses a fixed size window, shown here in black, for training. Subsequently, the method is tested against the data shown in orange.
On the other hand, the expanding window approach uses more and more training data, while keeping the testing window size fixed. The latter approach is particularly useful if there is a limited amount of data to work with.
It is also possible, and often best, to marry the two methods: start with the expanding window method and, when the window grows sufficiently large, switch to the sliding window method.Printable weekly planner with time slots
Many evaluation metrics have been proposed in this space, including absolute errors and percentage errorswhich have a few drawbacks.
- Quick med cards
- Opp report midwestern ontario
- Heavy lift drone design
- Bot traffic seo
- Headache after injecting ice
- La teoria delle descrizioni definite di bertrand russell
- The village of capriolo, municipality of ceccano (fr) lazio
- Bearing press fit tolerance
- How to uninstall qti logkit
- Dragon nest top tier dps 2019
- Diagram based vw polo 2004 wiring diagram completed
- Clock url
- Trapezoidal formula for volume in surveying
- Vespa immobilizer code
- Nano server 2019