Sensor Tech And IoT: Building The Intelligent Supply Chain
With the digitalization of supply chains, the first challenge was converting old processes into agile ones. The challenge now is not just organizing the data that has come from these new systems but understanding how IoT can further augment supply chain processes. In this article by Manufacturing Business Technology, find out how businesses can utilize IoT in their supply chain to get higher levels of output and ROI.
As the global data economy grows, more logistics companies have recognized that data, when sliced and diced the right way, can provide invaluable insights into their business. Business Insider Intelligence recently reported that investments to operationalize data are set to grow, estimating that $112 billion will be funneled into logistics and supply chain technology by 2019. The ability to ask and answer questions related to performance, risk management, and ROI can save logistics companies, whether in the luxury goods or pharmaceutical industries, time and money. The big question here is this: how can logistics companies transition their supply chain strategies from strict asset tracking and protection to a strategy that taps the power of the IoT to create a smarter supply chain?
Sensors and the IoT
To answer it, businesses must first understand how sensors power the IoT. By definition, a sensor is anything that translates the analog world into data. Without sensors, the IoT would simply not exist as we know it.
Logistics companies may not realize it, but they are sizable contributors to the IoT. Forbes and others have recently indicated that industrial and logistics companies stand to gain higher ROIs than any other industry by harnessing IoT and sensor data. Sensors transmit and collect data routinely on a continuous basis. This provides operations owners the ability to literally extend their eyes everywhere, “seeing” location, position, speed, temperature, pressure, lock status and 80+ more data points about all of their assets in near real-time.
Unfortunately, sensor data is typically raw and lacks business context. First it needs to be cleaned. Then it needs to be converted into human-readable formats. Next it needs to be combined with business context, rules and predictive models to transform data into actionable information that enables companies to make intelligent decisions to improve operations, performance and risk. Finally, all of this has to be easy (after all, logistics companies manage shipments, not data) and fast (as there is no point in sharing information long after you can use it for decision-making).
To illustrate this, here are three business issues solved by the intelligent supply chain, sensors and the IoT.
Case 1: Predicting and Optimizing Trip Time
Using sensor data to predict and optimize travel time provides a multitude of benefits. Predicting when shipments will be late allows logistics managers to immediately notify customers, rebook warehouse loaders or reschedule cross-docking operations. Analyzing data over time allows managers to optimize routes and schedules to reduce travel time, plan cross-dock plans with fewer disconnects and manage which drivers provide the best results by route, season, and time of day.
This solution combines data from sensor, supply chain logistics, weather, traffic and more and applying machine learning to detect repeatable patterns that managers can use to predict transit times and delays under a variety of circumstances. It presents this information in graphical interfaces that managers can access in seconds from any computer, tablet or smartphone, enabling them to make decisions wherever they are most useful.
We have helped companies use these solutions to reduce journey times up to 48 percent, saving thousands of dollars daily.
Case 2: Improving Asset Utilization
Maximizing freight capacity while minimizing fleet waste is another top concern. Typically this is done by combing through reports that are manually tallied and assembled across many levels of the organization. This is labor-intensive and error-prone.
On the other hand, sensors provide direct “eyes and ears” when assets are in use and when they are idle. It does this without making human time and effort to record information and tally it up. Combing sensor data with order information allows logistics managers to see the complete picture on asset use. For example, not only can managers see when their vehicles were planned to for deliveries, they can also actually know when the vehicle left the yard when it was actually moving and when it arrived inside the delivery location. Managers can also see patterns such as knowing that Driver A will always stop over on at a motel regardless of whether he leaves at 2pm or 4pm. As such, the carrier could use the truck longer before leaving without affecting the next delivery time.
By combining sensor and order data with information such as average fuel and truck lease costs, we can show customer revenue earned (or lost) per vehicle per day or week on-demand. We have shown companies how they can reduce fleet sizes by over 10% without reducing carrier capacity.
Case 3: Devising Efficient Pricing Models
Developing efficient pricing models is another critical issue. Sensors capture much information that can be used to measure the true cost of operations. This not only quantifies the direct cost of operations but also indirect costs such as time lost to risk, delay, and damaged goods.
In one case, a container rental company needed to develop a better pricing model because asset rental damage was eating away at its revenue. By combining sensor data, data on-the-spot human assessment, and order data we enabled the company to see which customers – and which types of customers – were responsible for the highest revenues and for returning assets late and with the most damage.
As a result, the rental company was able to change their pricing model to capture revenue more quickly and determine which types of customers were more likely to lead to higher revenues.
Key Business Takeaways
Combining sensor and supply chain management data creates a smarter, more intelligent supply chain that aids companies in solving a variety of business problems every day.
Sensors act as the eyes and ears of the organization, allowing managers to see what is really happening throughout their organization. Adding advanced analytics and machine learning converts this remotely captured raw data into end-to-end omniscience that empowers everyone from the warehouse manager to the chief operating officer can use to make better business decisions.
Jim Haughwout is Chief Technology Architect at Savi.
This article was written by Jim Haughwout from Manufacturing Business Technology and was legally licensed through the NewsCred publisher network. Please direct all licensing questions to legal@newscred.com.
