The better way to reduce shipping costs: Model and optimize logistics | solver

Higher prices of fuel and raw materials are impacting many businesses. Ongoing supply chain disruptions are crippling entire industries. Everyone is working to maintain the bottom line in the face of rising shipping costs. How can you help your organization weather these challenges? One powerful strategy for cutting costs is to use shipping cost optimization to create a more efficient supply chain. 

Let’s see how this approach can make a big difference for achieving your business goals. 

The business case to optimize logistics 

Logistics groups within organizations have historically underinvested in digital technology and have tended to rely on simple, cost-plus pricing. As McKinsey describes, many logistics and shipping/distribution teams could benefit from an analytics-based approach, with data-driven network models for decision making. For procurement teams, the past few years of shipping delays and disruptions have shown the importance of adjusting and optimizing supply chains on a regular basis for maximum efficiency.  

McKinsey outlines a few common business problems requiring a more advanced approach to optimizing logistics: 

• Extreme price fluctuations: During the pandemic, there was extreme uncertainty about pricing, and high volatility. For example, at one point, air freight prices from Asia to Europe doubled. 
• Volume surges: Recent higher-than-expected fluctuations in volume may suggest the need for a “buffer inventory” – but where and how much do you need?
• Higher demand for premium services: Companies have shown a higher willingness to pay extra for premium services like spot-cargo or low-carbon shipping. 

These are just a few of the potentially costly shipping problems that business leaders are up against. Especially in a time of heightened uncertainty and pricing volatility, the risks of guessing are costlier than ever. Instead, business leaders should leverage current data and employ advanced supply chain models to optimize their logistics networks. 

Why is supply chain optimization important?

Business leaders are familiar with the concept of “economies of scale,” but in the logistics world, sometimes the opposite happens. “Diseconomies of scale” occur in logistics when the per-unit cost of a load of freight increases with a higher volume of loads, on a given lane. These tight market conditions and spikes in demand can cause big disruptions, and can even cause supply chains to break down. 

Supply chain optimization can help your team reduce shipping costs and gain increased visibility into your logistics operations. A well-designed supply chain model with up-to-date shipping cost data can adjust automatically to bottlenecks or friction points, meet demand and still minimize costs. Optimized supply chains support informed decision making; saving time and boosting the bottom line. 

Never turn left: Lessons from UPS on how to optimize logistics 

One of the most prominent U.S. logistics companies, UPS, offers a surprising everyday insight into what “cutting costs” really means. As described in The Washington Post, UPS trucks almost never turn left. To be precise; the distinctive brown delivery trucks turn right 90% of the time. Why? 

Left turns are inherently more time-consuming and risky. Federal traffic safety data has found that left turns are involved in 53.1% of crossing-path collisions (compared to only 5.7% for right turns). Extra time spent signaling left at intersections and waiting for oncoming vehicles triggers traffic congestion and delays. By designing their routes to avoid left turns, UPS has been able to save millions of gallons of fuel, save time and reduce left-turn-related safety risks. 

This is just one example of how optimizing logistics can reduce costs. For companies like UPS, optimizing vehicle routes is central to their business. SVP-level managers are employed to focus solely on in-house optimization algorithms and solvers to maintain proprietary maps and traffic conditions. 

But other companies, of all sizes and industries, can improve their logistics optimization with a better approach to data analytics and modeling. 

How to get better business decisions with optimization modeling

Improved data-driven modeling can lead to lower shipping costs and a more profitable operation. Here are just a few examples of data-driven optimization models that can be used to help optimize logistics: 

• Transportation Models: Decide how many products to ship between factories and warehouses or to each end customer.
  
  With an explicit model of the network of possible shipping routes between factories, warehouses and end customers, as well as warehouse capacity and current stocks at each point, optimization can quickly uncover the best alternative ways to deliver products.
  
  When customer demand fluctuates unexpectedly or disruptions occur at supply points, re-running the optimization model with the revised data can quickly discover the best alternative routes, and again minimize costs in the new scenario.
   
• Partial Loading: Maximize the efficiency of your vehicle fleet by deciding which sizes/types of products should be loaded into each vehicle. Align product loading with your vehicle size limits to reduce wasted space.
  
  “Rules of thumb” for loading vehicles can be fine if conditions are constant and predictable. But change is the new norm, whether due to fluctuating demand, or new or substitute products. With an explicit model of vehicle space, product sizes and shapes, and other constraints, the computer can adapt to last-minute changes and pinpoint “non-obvious,” more space-efficient loading patterns.
   
• Facility Location: Decide which factories or warehouses (if any) to open or close in order to reduce shipping costs and overall fixed operating costs.
  
  Facility decisions are infrequent, but often involve many millions of dollars in costs with long-term consequences for a company’s ability to meet delivery and time requirements. With an appropriate model, you can explore all possible combinations of facility locations, along with many scenarios for customer demand and distance-dependent shipping costs. 
   
• Production / Transportation Model: Decide how many products should be produced in each factory, and how many products to ship to various warehouses and customers. 
  
  If you can decide which and how many products to make in different factories at different locations, you know warehouse locations and capacities, and you can estimate customer demand at many different endpoints, it’s possible to build an integrated optimization model. 
  
  An integrated optimization model takes into account the type and number of products manufactured at each factory, warehouse locations and capacities and customer demand. Using an integrated model, business leaders can seek answers to “what if” scenarios using different estimates for demand and/or shipping costs. Solver will find the best combination of production and shipping conditions that minimizes the bottom line. Stochastic optimization models yield robust capabilities meeting uncertain demand across a wide range of future scenarios.

All of these models reduce shipping costs by aligning factory supply with warehouse demand, while managing capacity of the logistics network. 

Ready to optimize logistics? Discover how.

Optimizing logistics doesn’t have to involve a big capital investment or a time-consuming transformation of your business operations. A Deloitte study found that one company was able to save $1.5 billion by using analytics to redesign and optimize their product supply network. 

Better use of analytics for your logistics network can help you reduce costs, save time, and manage risks. And that’s not all: Check out this example from Forbes: in Japan, Toyota partnered with local nonprofits, hospitals, and emergency responders to create a new service to transport vehicle crash victims to the hospital as quickly and efficiently as possible. 

Toyota’s new partnership uses vehicle crash data and advanced analytics to identify the optimal route to the nearest hospital. Using traffic and route analytics, emergency responders can make real-time decisions on whether to send an ambulance or helicopter, based on the severity of the crash. After a severe traffic crash, every second counts. Optimizing logistics is not just a matter of reducing shipping costs. Sometimes, analytics-based logistics decision making can literally save lives. 

Ready to see how optimization models with current data can improve your logistics performance? Sign up for a FREE 15-day trial of Analytic Solver®, our all-in-one solution for Optimization, Simulation and Data Mining.