Tom Moore shares how ProvisionAi provides seamless integration of planning and execution
I was born and raised in New Zealand, studied mathematics as both an undergrad and postgrad and then went on to get a degree in operations research. I decided to seek my fortune in America and started work in an industrial company. I was promoted very quickly to positions that nobody my age really should have been, running truck fleets and warehouses, implementing an entire ERP system and managing a sales office for a short period of time. It really was a baptism of fire and I still have the marks on my back to prove it,” laughs Tom Moore, CEO and Founder, ProvisionAi.
“I was then approached by a different firm and went on to work in supply chain consulting. After a while and having realized the disparity between the revenue I was generating and the salary I was earning, I decided to go it alone. One of my first clients was Procter & Gamble (P&G). Historically, they shipped truckloads of a single product, such as liquid detergent, which was a relatively straightforward process. When they came to us, they were trying to work out how best to ship a range of products requiring smaller quantities of each item: a few cases necessitating a huge increase in order selection. They were looking for help with automating the thinking behind their case picking. For many years, we worked with P&G under a nondisclosure order. Alongside some of my university friends, we created a software system that solved their problem and enabled the most effective use of transportation capacity,” he explains.
Maximizing efficiency
Today, AutoO2, ProvisionAi’s optimization solution leverages artificial intelligence to optimize shipments with over 300 parameters, including axle weight, stacking, and both product and customer-specific loading rules. The solution can be customized further to align with existing systems. It uses advanced technology to create load plans to fill trucks fuller, as over 90 percent of all trucks are not filled to capacity. This then reduces the number of trucks needed, leading to better sustainability and reducing cost. Using advanced mathematics, AutoO2 increases the truck payload by five-to-ten percent and removes unneeded journeys from the road.
The ProvisionAi platform helps organizations smooth their transportation workflows. Using AI, its complementary solution, LevelLoad, looks out to the next 30 days to see what replenishment requirements must be shipped and which ones can wait. Then it evens out the flows so that the number of required trucks doesn’t fluctuate dramatically from one day to the next. ProvisionAi automatically gets the right products on the right number of trucks at the right time to increase service levels and maximize network efficiency.
“Our founding team of supply chain veterans and machine learning experts has created one of the most sophisticated suites of optimization tools in the supply chain world. We took 88,000 trucks off the road last year, and we’re doing that from Australia to Poland. We continually refine our technology. It’s not just about packing a truck to capacity, it’s also about making sure that when the load reaches the shipping dock, the guys picking it can achieve what was planned. Likewise, when the product reaches the customer, it can’t be damaged. Along the way, there are additional legal and safety complexities that must also be managed effectively in accordance with the unique requirements of each individual country,” continues Tom.
Groundbreaking solutions
LevelLoad ensures deployment plans respect supply chain constraints, such as high customer service, transportation availability and warehouse capacity. It uses a combination of linear programming and reinforcement learning to create more efficient schedules across a transportation network.
“We’ve patented this methodology, and the beauty of today’s operating environment is that we can handle vast quantities of data needed to manage such a complex task. We have access to data from ERP, supply management, and transportation and warehouse management systems, and use AI and reinforcement learning, to pull together an incredibly effective and quick solution that smooths out volatilities. Ironing out volatilities enables savings to be made. Carriers build inefficiencies into their rates, so by removing those inefficiencies, more favorable rates can be negotiated. Huge transportation savings have been made, quickly recovering implementation costs. This solution is loved by warehouse operators, workers, and carriers; all the people who reap the benefits of its efficiencies.
“Organizations, such as Kimberly-Clark, are using our solution and consequently saving millions. Being able to use network freight demand to match preferred carriers and distribution center capacity has been a real game-changer for Kimberly-Clark. LevelLoad helps to optimize operational efficiency, reduce costs, and improve customer satisfaction by enabling better planning and execution. The LevelLoad transportation replenishment schedule considers capacity constraints and allows early tendering, smoothing out the movements across their network. The solution performs daily optimization for the next 30 days and generates a globally optimized replenishment transportation schedule for the month, recommending early trucking capacity reservations across the entire network. This approach creates a smooth, optimized flow of products throughout the supply chain, saving money while increasing customer-order fill rates and timeliness.”
ProvisionAi’s groundbreaking solutions, LevelLoad and AutoO2, have redefined transportation scheduling and load building, leading to enhanced customer service, optimized freight carrier utilization, and substantial cost savings. These patented innovations have been pivotal in achieving operational excellence and sustainability for numerous renowned clients.
“It really is a win-win,” Tom concludes. “User acceptance is normally the problem when introducing new systems. For us, we’re seeing the opposite; customers are asking us to roll out our solutions and expand to other areas quickly, which really is a high-class problem to have.”