Infosys’ blog on industry solutions, trends, business process transformation and global implementation in Oracle.

« Have you planned the testing of custom reports in your Upgrade Project? | Main | Supply chain visibility »

Optimization overdrive

Eventually, every business must optimize its processes. Every organization has goals - be it improving revenues, increasing its customer base, or even sustaining itself in a competitive environment. Organizations incentivize their employees to achieve their annual goals and it often works out very well. But in certain cases, such as the logistics industry, where multiple organizations collaborate and compete at the same time, and where forward integration is always looming right around the corner, the simplified focus of improving profits year-on-year may not be the best approach.

Forward Integration is ruthlessly executed by Freight Forwarders and Custom Brokers whereby they add a long list of logistics services to their portfolio, eroding the customer base from existing logistics players in the market.

An order placed by the end customer is typically subjected to multiple iterations of optimization by the manufacturer, carrier, third-party logistics (3PL) service provider, freight forwarder, etc. This article presents the downside of optimization in the logistics industry, in favor of a single meta-optimization.

Retailer's perspective

Can companies collaborate to optimize total supply chain costs? Can a company forego its immediate margins for a sustainable future?

Let us consider a supermarket in a country like Australia where land-based transportation is ubiquitous. This supermarket has its own fleet but primarily depends on 3PL service providers for most of its transportation needs. This supermarket, like most players in this segment, thrives on stocking fresh consumables and moving the items off the shelves in the quickest manner possible. To ensure this, two things need to happen:

  1.        The fresh food products need to be delivered from the vendors to the supermarkets as quickly and as efficiently as possible. Any delays from the day the food was made available by the vendor to the day it arrives in the supermarket can have significant repercussions
  2.       For the retailers, buying 3PL's transportation services should be beneficial in the long run over using their own fleet

Its transportation management system enables the supermarket to place orders, optimize, track the movement of goods, and so forth. Goods move from the vendor's location to the distribution center. This comprises the first leg of the movement. Here, workers reassemble the cartons / pallets into smaller trucks to deliver goods to all the supermarkets in a particular zone. This second leg is usually executed in the form of a multi-stop model.

It may not be obvious at first, but it is really challenging to optimize costs on first leg movement because transportation is carried out by a 3PL. This 3PL consolidates all the orders they receive from multiple customers (supermarkets, manufacturers, and other retailers) in a way that is beneficial to them. The 3PL users initiate their load plans / shipments on top of whatever optimization was already made at the customer's site.

So, for instance, the supermarket in question has a 100 orders on a given day, which are to be picked up from 10 pickup locations and be delivered to their DC within three days. The supermarket's TMS takes these 100 orders and optimizes them based on:

  1.         The distance between pickup locations
  2.       The requested delivery date of individual orders
  3.       The truck capacity in the available lanes

Based on these factors, the supermarket's TMS arrives at say, four load plans / shipments. These are transmitted via Electronic Data Interchange (EDI) to the 3PL's TMS system.

The 3PL TMS system receives multiple load plans / shipments from multiple customers. This system also receives unplanned orders placed by smaller customers who don't have their own TMS systems yet.

The 3PL TMS performs an additional consolidation. It considers the very same factors that the upstream TMS has already considered, thereby disputing the original plans. The following scenarios are then possible:

  1.         The original load plan / shipment received from the upstream system considers a capacity limit of, say 4000 cubic meters, in a certain lane and the volume utilization achieved was around 82 percent. The downstream TMS can achieve a better volume utilization by upgrading a smaller truck to a larger truck and consolidating several other orders / shipments, effectively achieving a volume utilization of 98 percent. Of course, this is subject to slight changes to the original plan received from the supermarket
  2.      3PLs make the delivery to the doorstep, however, the supermarket only negotiates for the delivery to be made to their DCs. After the picking and packing is done, the supermarket has to buy the 3PL services a second time for the second leg of the movement. Instead of two shipments for the same order, a 3PL can make one delivery for a single price, thereby reducing the cost

3PL's perspective

Let us consider a 3PL in the same region as above. They would have their own fleet but a majority of their business operations are based on established contracts with multiple carriers. Some carriers specialize in providing line haul services between states while other carriers are more focused on regional deliveries.

A 3PL's buying behavior could be best described as follows:

  1.       Buy the whole truckload (TL) space if it is between major cities like Sydney and Melbourne, Perth and Brisbane, etc. This is because volumes are high and relatively stable, albeit subject to seasonality issues from time to time. Additional expenses such as fuel surcharge and accessorial costs are levied on the whole truck. Individual cartons, pallets, and consignments that make up the truck do not influence the cost of a TL. The 3PL has to make sure they maintain healthy volume utilizations on these trucks. The profit and margin that the 3PL makes on each carton / pallet / consignment is directly proportional to the truck's volume / weight utilizations. So, hiring a full truckload is only practiced on lanes where the 3PL has huge volumes
  2.      Buy less than truckload (LTL): This contract is based on the consignment's volume, weight, or other dimensions. The carrier would levy an additional fuel surcharge on each consignment. The total truck's utilization does not influence the profit and margin the 3PL makes on each carton / pallet / consignment. This is adopted for home deliveries where the customer is more than happy to upgrade to an 'expedited' service or an 'air overnight' service in place of a 'general' service. It is also useful in return deliveries where products purchased earlier are being returned to the manufacturer such as when a mobile phone that is ordered online arrives home with a manufacturing defect. If it is left to the 3PL, they would choose to negotiate only LTL rates for these and many other scenarios within the city limits. But some carriers that operate in remote locations demand fresh contracts that are neither TL nor LTL
  3.      Hourly hire: Here, carrier charges are truck-agnostic, so to speak. The 3PL will be charged only after the consignment is delivered to the recipient. It is at this point that the number of hours spent in delivering the consignment are recorded and relayed back to the 3PL. Most of this is automated; for instance, the driver of the truck opens up an app to get the recipient's signature. This app records the milestone achieved (proof of delivery) and electronically submits it to the 3PL. Since all the milestones are tracked, the 3PL's TMS can ascertain the number of hours spent on the job

Let us look at 3PLs in Australia. The 3PL's TMS receives thousands of orders every day. These orders are planned based on fixed routes that make use of their depot locations across the country. The depot locations serve as cross-dock facilities for line haul movements, typically carried out by TL carriers. Optimization is ruled out in line haul scenarios owing to the way contracts are negotiated with these carriers. The first mile and last mile carriers usually charge the 3PL based on each consignment's weight and volume.

With this in mind, let us look at their planning / optimization model. On a given day, let us consider that a 1,000 orders were placed by different customers, with each order also bearing a carrier and service nominated on them. The nominated service is used by the 3PL as a reference point for charging their customers. The 3PL's TMS can only optimize the first leg movement and the line haul movement. The last mile cannot be optimized because the customer has already nominated a carrier and the service.

Since the last mile cannot be optimized, the 3PL creates a load plan / shipment based on the requested delivery date (RDD) alone. So, the TMS segments all available orders and delegates different shipments based on what has to be delivered in N days, N+1 days, N+2 days, etc.

It is not the most efficient of planning scenarios. These load plans are then electronically submitted to the last mile carrier. The last mile carrier takes these shipments and consolidates them with other shipments / orders that it receives from other customers. So the downstream carrier's TMS essentially discards the planning that was performed in the upstream 3PL's TMS system.

In this case, the following scenarios are possible:

  1.        The 3PL's TMS creates three shipments to be delivered today, tomorrow, and the day after tomorrow respectively. These shipments have the earliest start time of yesterday, today, and tomorrow respectively.  When the carrier receives these shipments, and their TMS looks up available trucks to optimize, it is limited due to the time window sanctions imposed by the 3PL's TMS system
  2.      The carriers can benefit by maximizing their volume utilization. Ideally, a carrier should be allowed to switch between small and large trucks based on the delivery time windows of the original orders. But the original orders and the original time windows are not visible to the carrier. The carrier's TMS is fed processed data from the 3PL's system

Residual cargo carrier's perspective

Containerization has evolved from a novel idea to a global phenomenon over the last century. Container freight stations may be owned by carriers, the government, or a third party warehousing player. When it comes to ports, the port authority of that country plays a crucial role in determining the inbound and outbound volumes.

In the international movement of goods, the sea ports, the respective container freight station (CFS) locations, the availability of residual cargo carriers, and the distance between them influences the extent of optimization. For instance, a shipment from Shanghai to New York can be achieved in:

  1. A single-leg voyage all the way from the port of loading to the port of discharge
  2. Multi-leg voyage with the first cross-docking at Singapore and onwards

Ships usually have dedicated routes with the vessel schedules being fixed. There are simply far too many documents to fill out at each port and the volumes are just too high to be able to carry out a multi-pickup or a multi-delivery mode of transportation. There are also cutoff times for goods to arrive at all the ports, beyond which goods may not be considered for documentation purposes. For these reasons, all ocean-based modes are direct shipments, from port to port.

Let us consider the CFS in Singapore. Singapore is a cross-dock facility for many carriers. For instance, if a retailer in North America is planning to stock the goods before Christmas, they may place their orders with their supplier in Shanghai as follows:

  1.         Male garments with a delivery time window of four months
  2.       Female and kids garments with a delivery time window of three months
  3.       All footwear with a delivery time window of three months
  4.       Promotional products with a delivery time window of one month (so that the products can be displayed months before release)

Now, let us add similar orders with similar time windows from Beijing to a Latin American retailer to this mix.

Since the routes are fixed between Shanghai and North America, and between Beijing and Latin America, the only scope for optimization is in cross-docking at the CFS in Singapore. When the carrier's TMS receives the orders listed above, they would be consolidated with other orders into 20FT or 40FT containers on the next outbound vessels from the respective ports. Of course, the urgency of sending an order outbound depends on the requested delivery dates by the customer. Having said that, the first leg movement's objective would be container capacity utilization. If there is an outbound vessel that can carry about 80 percent of all the orders placed, they will be shipped right away, without considering the delivery dates. These shipments would reach Singapore and hibernate in the CFS based on the urgency, as dictated by the delivery time windows.

When the next vessel is scheduled to leave after, say a month, the residual cargo from earlier orders would be shipped in it. But if all the shipments can only make up to 10 percent of the ship's capacity, the original carrier would outsource it to another carrier. This outsourced carrier would take all such residual cargoes and reach the Singapore CFS where the goods are deconsolidated to be warehoused / shipped for the future.

Herein lies the problem. The original cargo carrier would have his own dedicated ships and would try to optimize in order to achieve the highest container utilization. So, the original carrier would wait until a cutoff time in order to give himself a decent chance at accumulating multiple orders into a single vessel. After the said cutoff time, the original carrier would make the decision, either to ship it himself or to outsource. During this time, the residual cargo carrier is also giving himself a decent chance to consolidate shipments.

Thus, the following scenarios are possible:

  1.        The residual cargo carrier has enough shipments with him to load the next vessel before the original cargo carrier can place his outsource orders. Now, the original carrier has to wait for the next vessel, and if it proves to be too late, he may have to consider an aerial route
  2.       The residual cargo carrier commits to a vessel schedule but does not have enough shipments to load on board

Conclusion

This brings us to the question of seamless optimization. Can the logistics industry evolve to overcome the above predicaments? Can multiple TMS systems interact and collaborate to benefit everyone involved? Can upstream and downstream systems intelligently predict the optimization of each other, without having to work in isolation?

In the case of the retailer in Australia, instead of committing to load plans / shipments created in their TMS system, they could submit a draft of their shipment plans to the downstream TMS. The downstream TMS can run a consolidation plan on all the draft shipment plans received from multiple customers. After a certain period of time, profit, margin, and other key performance indicators (KPI) can be compared between consolidation plans executed on draft shipments and the same plans executed on committed shipments for better planning.

In the case of the 3PL, instead of sharing the processed shipment plans with the carrier, raw orders can be submitted to the carrier. The carrier can run a draft iteration of optimization on the orders received from the 3PL for a given month and submit the draft shipment plans back to the 3PL. The 3PL can review this output. The original orders usually drop in with constraints such as their compatibility with other stock keeping units (SKU), whether to refrigerate, to be handled carefully for fragility, etc. The original orders also have the delivery time window, nominated carrier, and type of service specified on them. If the carrier's submission of the draft optimization satisfies all these conditions, the 3PL can accept the draft shipments or make amends to these draft shipments before finalizing. Initially, this may have to be handled by transportation managers from either side. But as the interface evolves, the carrier's TMS can be made intelligent to track the acceptance ratio of all the draft plans that it submits to a certain 3PL. These acceptance indicators can be used in predicting the acceptance from the upstream or downstream TMS', and adjusting the plans accordingly to gain better acceptance and so forth.

The residual cargo player can request submissions from all the original carriers in the area. A constant feed of all the residual cargoes can be submitted via an automated interface between all the carriers. The residual cargo carrier can execute an iteration of optimization at the end of every day to verify container capacity utilizations. The system can track the percentage of utilization of containers and the next outbound vessel. A notification can be sent to all the carriers once the residual cargo carrier's iteration yield exceeds a certain percentage. This will enable all the original carriers in the area to plan and execute their shipments in a much more efficient manner.

TMS systems working in isolation are inferior to TMS systems collaborating, thereby achieving an all-encompassing system. An omniscient TMS system like this can iron out inconsistencies over a period of time, rule out uncertainties, and may even prove to be resilient in times of unforgiving global economic changes.

To know more on optimizing logistics, please meet us at the Infosys Booth during the OTM SIG APAC 2016 Conference (Singapore) and we shall be delighted to showcase our solutions.

Written by: Kranthi Sagar Askani

Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)

Please key in the two words you see in the box to validate your identity as an authentic user and reduce spam.

Subscribe to this blog's feed

Follow us on

Blogger Profiles