{"id":252354,"date":"2025-10-21T16:12:48","date_gmt":"2025-10-21T08:12:48","guid":{"rendered":"https:\/\/www.grab.com\/sg\/?post_type=editorial&p=252354"},"modified":"2025-10-21T16:13:07","modified_gmt":"2025-10-21T08:13:07","slug":"were-making-the-power-of-forecasting-available-to-more-grabbers","status":"publish","type":"editorial","link":"https:\/\/www.grab.com\/sg\/inside-grab\/stories\/were-making-the-power-of-forecasting-available-to-more-grabbers\/","title":{"rendered":"We\u2019re making the power of forecasting available to more Grabbers"},"content":{"rendered":"\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t

Imagine it\u2019s rush hour when suddenly, a heavy downpour begins. Quick shifts in weather are common in Southeast Asia. Demand for rides surges, but Grab still manages to match passengers with drivers quickly, keeping waiting times reasonably short. Behind that smooth experience lies a lot of careful planning\u2014made possible with one of the most powerful tools in our toolbox: forecasting.<\/p>

Forecasting helps us predict what\u2019s likely to happen in the future. Whether that\u2019s how many rides will be requested in an area, how driver supply will shift with weather and time of day, or even how our systems might behave under different conditions.\u00a0<\/p>

Many industries rely on it: airlines forecast demand to optimise ticket prices, supermarkets forecast sales to manage inventory, and logistics companies forecast traffic to plan deliveries. At Grab, forecasting underpins everything from driver-partner incentives to ride pricing. As a Grab user or driver-partner, you\u2019ve likely benefited from our enhanced forecasting models without being aware of how it works behind the scenes.<\/p>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\t\"\"\t\t\t\t\t\t\t\t\t\t\t
Forecasting plays a big role in many industries. For example in shipping, forecasting is essential to help anticipate demand for containers along various routes.<\/figcaption>\n\t\t\t\t\t\t\t\t\t\t<\/figure>\n\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t

However, sophisticated forecasting is a complex task. Not every Grab team had the resources or expertise to set up forecasting pipelines from scratch. We needed a solution that simplified the process, while delivering high-quality results. This is what led to the development of Spyce.<\/p>

The birth of Spyce<\/h5>

Spyce is an internal tool we created. It wraps multiple forecasting packages together, lowering the barrier to entry making advanced forecasting accessible to any team at Grab.<\/p>

It came about as we were working on improving forecasting for a specific problem: Dynamic pricing. We were trying to understand how best to make subtle price adjustments to Grab rides\u2014depending on the supply and demand situation\u2014 with the goal to balance out the marketplace for optimal service, considering real-time changes on the ground.<\/p>

[<\/span>Also read: Why and how do ride fares on Grab change with time?<\/span><\/a>]<\/span><\/p>

As we pooled our resources to solve this problem, we noticed a pattern. Actually, many forecasting tasks across Grab\u2019s divisions and teams followed the same steps: gathering data, selecting the right forecasting models, defining metrics, and then evaluating performance. <\/p>\n

With Spyce, we created a tool that accelerates the process of building a forecasting pipeline through automation. For example, it will automatically select the right model for the task, and works with the standardised metrics already available via other internal tools Grabbers use.<\/p>\n

Non-technical users don’t have to install or write any code to work with Spyce. They can just send a request to forecast a given metric to an internal service and receive a result. This service will take care of all the work related to model selection and data preparation. The Spyce package can also easily integrate with AI agents (e.g. chat bots) that teams can then use to generate forecasts on their own.<\/p>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\t\"\"\t\t\t\t\t\t\t\t\t\t\t
Spyce can be used by all teams across Grab, even those with less technical proficiency due to its easy integration with other popular internal Grab tools.<\/figcaption>\n\t\t\t\t\t\t\t\t\t\t<\/figure>\n\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t

In short, it helps any team get reliable forecasts faster, with less setup and guesswork. It reduces usual data-science work from weeks to hours, and allows teams to focus on the business problems and outcomes we\u2019re trying to solve for, rather than spending time on setting up the forecasting models.<\/span><\/p>

As for the name, Spyce is inspired by \u201cthe spice,\u201d from the science fiction universe <\/span>Dune<\/span><\/i>. It\u2019s a mysterious substance that helps those who ingest it gain foresight and heightened awareness.<\/span><\/p>

The power of Spyce<\/b><\/h5>

Spyce started as a collaboration with the Fulfilment team, but thanks to support from the tech infrastructure team and others, it\u2019s now expanding its reach across Grab. Any team can integrate Spyce into their workflows to build forecasts that guide smarter decisions.<\/span><\/p>

Today, Spyce powers forecasting in a growing number of use cases across Grab, including:<\/span><\/p>

  • Advanced Booking Nudges: When we predict low fulfillment rates in certain areas, the app can nudge users to schedule an advance booking\u2014improving their chances of getting a ride when they need it.<\/span>
    <\/span><\/li><\/ul>
    • Supply Planning: We can forecast future ride demand more accurately and nudge drivers to close supply gaps.<\/span>
      <\/b><\/li><\/ul>
      • Smart Alerts:<\/span> Spyce supports anomaly detection in key business metrics, triggering automated alerts when incoming data begins to deviate from a forecasted time series. This empowers teams to respond early, with proactive interventions.<\/span>

        <\/b><\/li><\/ul>

        Our next steps include enhancing Spyce\u2019s automation, improving its scalability, and integrating it even more deeply across Grab\u2019s systems.<\/span><\/p>

        At the end of the day, all this innovation serves one purpose: to make every Grab experience smoother, more reliable, and more delightful\u2014whether you\u2019re a driver-partner, a passenger, or someone waiting for their meal to arrive right on time.<\/span><\/p>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t

        \n\t\t\t\t\t\t
        \n\t\t\t\t\t
        \n\t\t\t
        \n\t\t\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<\/div>\n\t\t","protected":false},"parent":180237,"menu_order":0,"template":"grab21-default","acf":[],"_links":{"self":[{"href":"https:\/\/www.grab.com\/sg\/wp-json\/wp\/v2\/editorial\/252354"}],"collection":[{"href":"https:\/\/www.grab.com\/sg\/wp-json\/wp\/v2\/editorial"}],"about":[{"href":"https:\/\/www.grab.com\/sg\/wp-json\/wp\/v2\/types\/editorial"}],"version-history":[{"count":50,"href":"https:\/\/www.grab.com\/sg\/wp-json\/wp\/v2\/editorial\/252354\/revisions"}],"predecessor-version":[{"id":252411,"href":"https:\/\/www.grab.com\/sg\/wp-json\/wp\/v2\/editorial\/252354\/revisions\/252411"}],"up":[{"embeddable":true,"href":"https:\/\/www.grab.com\/sg\/wp-json\/wp\/v2\/editorial\/180237"}],"wp:attachment":[{"href":"https:\/\/www.grab.com\/sg\/wp-json\/wp\/v2\/media?parent=252354"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}