Ambi Robotics Prepares Customers for Peak Season with New Systems and Support

The AmbiSort B Series can streamline final-mile delivery operations with automated induction, scanning, and end-to-end sortation, according to Ambi Robotics.

As Black Friday and the peak season approach, logistics providers are preparing for a surge in demand. Each year, these providers could see a 25% to 40% increase in throughput, according to Stephen McKinley, co-founder and COO of Ambi Robotics. This year, Ambi expects its customers to experience a 9% to 10% increase compared with last year's peak season.

McKinley told The Robot Report that despite economic uncertainties, volume continues to ramp and e-commerce spending remains high. This puts pressure on clients to perform at all their sites, and Ambi feels that pressure too because its systems are core to their operations.

To prepare, Ambi is installing new systems at eight facilities before peak season starts. Most of these new sites use a new version of AmbiSort, which McKinley described as a new variation rather than a new product. The company is excited to support these systems through peak season and wants everything to go perfectly for its customers.

The Berkeley, California-based company is also ensuring its team members are readily available for logistics customers. McKinley said the company moves into a more reactive state, with about half the team traveling to different sites to check on operations, while the other half is ready to jump in and solve issues as they arise.

While some customers plan to add more systems for peak season, there are other ways to increase throughput. McKinley noted that customers often add more shifts to existing operations or bring on more people to work alongside the systems, and Ambi helps with training. The company offers more training right before peak season to ensure operators know how to use the systems effectively.

Properly training operators can make a big difference in system performance. McKinley said Ambi continuously tracks system performance throughout the year. Performance depends on the inherent quality of the machinery, which doesn't fluctuate much, but also on how customers interact with the machine, how attentive they are, and how well-trained the operators are. Ambi steps in to make sure everyone knows what to do and has the right mindset.

Reducing downtime is especially important during peak season. To achieve this, Ambi forward-deploys spare parts to different sites. Normally, the company keeps spare inventory at its Berkeley headquarters and ships as needed. That inventory is also used to build new systems. During peak seasons, Ambi brings in more spares and deploys them to specific locations on the East Coast that are hubs of customer activity. For example, Cincinnati is a central zone for many of Ambi's installs in the Midwest. The company even places spare computers or other high-critical items at each site to carry more inventory at customer locations during peak season.

When it comes to rolling out improvements to its AI systems, Ambi takes a conservative approach. McKinley said Ambi has been a physical AI company since 2018 and knows how important it is to have systems performing at predictable levels. Therefore, the systems do not self-improve over time. Instead, as Ambi generates more data from its systems, it uses that data to retrain its systems and then processes the improvement through an A/B test in a controlled environment before rolling it out to the rest of the field.

Building trust with customers is just as important as improving technology. McKinley explained that the people answering the support hotline are the same people who go to facilities to do training and make fixes. This builds a layer of trust with operators, on-site maintenance staff, and management. When customers call, they recognize the voice on the other end, and that person is familiar with them. McKinley added that this trust has helped significantly and, in his opinion, has been the most important factor for large customers adopting physical AI in the real world.

Instead of using AI to directly interact with customers, Ambi has developed an AI system to help its employees solve problems more quickly. McKinley said the company has been collecting data from support tickets and feeding it into its own AI model to assist in support. With a very lean team, Ambi now uses AI assistants to keep track of performance on sites, predict when things might go wrong, and help the support organization.

The Robot Report has opened submissions for its TRR50 list and encourages innovators to submit their entries.

This article was originally published on The Robot Report.