Packaging on Demand Solutions
CIOREVIEW >> Consumer Packaged Goods >>

Packaging on Demand Solutions

Mark Morrison, Chief Information Security Officer, OCC


Pallet packaging decisions are often an overlooked area of a warehouse operation. However, the pallet remains the primary storage medium in warehouse operations, decisions such as how to orient case patterns on a pallet can lead to significant storage utilization increases. We have developed three models that target getting more pallets into a warehouse. All of these models work within current infrastructure and do not come with any capital outlay, just changes to cases orientations and rack heights. Case studies have shown storage utilization increases of up to 20 percent can be realized, leading to getting out of expensive outside storage and trailer demurrage charges. The fourth model takes a look at carton utilization in a warehouse and addresses the problem of packaging orders into cartons that are too large for their contents, this leads to additional dimensional weight charges from the parcel carriers. Since shipping is often 2X fulfillment costs, this can be very significant. Our carton optimization model (US patent pending), identifies the best preconfigured carton array that should be used to match the order profile. A case study at a large fulfillment operation yielded a 12 percent reduction in shipping charges. Cumulatively we call these solutions - POD (Packaging On Demand)


• Pattern Optimization: Pallet case patterns usually follow simple configurations. In one example, each layer of a pallet has six cases along the one side of the pallet repeating three times. This particular configuration would yield 18 cases per layer, but leaves unused space on the pallet footprint. A more complex pattern increases the cases per layer to 21 cases with no cases overhanging from the pallet. Note, case overhang may be acceptable in certain situations and thus utilization could increase further. Across four layers on a pallet this is a 12 (3 cases per layer X 4 layers) case difference and increases storage utilization by 15 percent. Note, this new case pattern is not intuitive and calculated with a complex algorithm. However, once set could be usually implemented at the end of the manufacturing process into the automated palletizer’s configuration template, leading to better utilization in both storage and transport.

• Pallet Stackability: Pallets are often stored on the floor and stacked on top of each-other. The ability to stack pallets safely depends on several factors: case orientation vertical edges, storage time, case weight, humidity, materials, etc. Stack height decisions are often arbitrary and sometimes done through trial and error. In numerous instances, pallet stack heights can be increased simply by changing the case orientation on a layer to create more vertical edges. Increasing stack heights from two to three represents a 50 percent increase in storage utilization and a significant space savings. Product storage guidelines within a company are often universal, but sometimes it is possible to stack pallets higher in a low humidity region for example a warehouse in Denver or Phoenix may have a greater stack height potential versus a warehouse in a high humidity zone like Atlanta.

• Rack Height Optimization: Rack heights are often set to accommodate a wide range of pallets, warehouses that have pallets of various heights often have poor cubic utilization, as small pallets end up in large rack positions. For example a warehouse that short 35” and tall 70” pallets, may set all their rack heights at 80” to accommodate both, creating fewer overall slots. Our optimization model looks at the mix of rack heights to the inventory profile. In this particular example, two additional 40” slots could be creased for every 80” slot that is removed thus increasing the number of rack positions and better fitting the inventory profile.

• Carton Optimization: In e-fulfillment and other unit pick operations, orders are often placed into cartons for shipping. The size of the carton used is often somewhat arbitrary and depends on a combination of what is available, packer judgment and WMS guidance. With the recent changes in how parcel carriers charge for dimensional weight this relaxed approach to choosing carton sizes often leads to additional shipping charges. Our model looks at the order profile and matches the best preconfigured array of cartons to maximize carton utilization. Results show a decrease of >10 percent in shipping charges can be realized through intelligent use of carton selection. Also our model serves as a good gauge of when to use ‘made-to-order’ packaging to increase carton utilization even further. Could become an Idea module for cartonization.

The Math

POD runs on advanced analytical algorithms to solve the aforementioned issues. Following points highlight the methodology used for each solution:

• Pattern Optimization: is developed using a combination of four geometric constructive heuristics; each heuristic provides optimal case orientations based on defined initial state and corresponding build strategy. Finally, the solution which is best among the four is chosen. Additionally, we also have a MIP to solve smaller problems (i.e. larger case sizes).

• Pallet Stackability: is additional simulation based heuristic which takes output from Pattern Optimization. Then using formulae for aspects like number of layers, humidity, case orientation etc. it provides the optimal pallet stackability for different case sizes.

• Rack Height Optimization: is combination of a meta-heuristic based on Genetic Algorithms (GA) which can find optimal rack heights for a large warehouse. Additionally, we developed an MIP for smaller and mid-size warehouses.

• Carton Optimization: is a hierarchical solution which has three tools. Tool 1 is a 3D cubing algorithm which used a novel methodology called MLAFF (Modified Largest Area Fit First). Tool 2 is a clustering heuristic which is based on minimizing dimensional gap between potential solutions. Finally, tool 3 is a GA based algorithm which generates optimal carton set using dimensional, geographic and cost constraints.

Case Study

In May of 2018 several of two of these tools were utilized at a large general merchandise retailer’s regional distribution center, which as at capacity with over 400 trailers constantly sitting in yard for an average of five days prior to being received. This represented 14 percent of overall inventory. Our analysis recommended changes to rack heights and increasing cases per pallet (primarily for imported product where retailer had complete control over pallet configurations).

Costs Eliminated

• Trailer demurrage charges: $3.2M (40K trailers/year * $20/ day * 4 days)

• Additional Shuttle = $800K (40K trailers/year * $20/shuttle)

• Total =$4.0M


Storage utilization is often overlooked in warehouses. Analytical models can greatly increase utilization by:

• Increasing cases per pallet

• Increasing pallet stack height

• Changing rack height configurations to allow for more pallet positions

Increases in storage utilization can lead to significant savings in avoiding outside storage and demurrage charges, and also significantly reduce inbound transport costs. The Carton optimization model, which addresses dimensional weight charges, perhaps has the largest potential for savings in this booming ecommerce environment as significant savings can be realized by optimizing the array of cartons to fit a shipper’s profile. All together these models combine synergistically to drive significant savings, warehousing and parcel costs both exceed $100B each.

Read Also

Cyber Insurance: Evolving Threats Demand Cutting-Edge Coverage

David M. Finz, First Vice President, Alliant Insurance Services

Transformation to Fit an Agile Future

Maria Luisa Inofre, CHRO at AboitizPower Human Resources

Gender and Racial Diversity in Australia's Senior Technology Leadership

Subha Chari, Head of Digital Product Delivery, LendLease

Impact of Digital Transformation in Retail Space

Robert Sjostrom, President Global Operational Services, Essity

Challenges Over The Past 18 Months

Marc Ashworth, Chief Information Security Officer, First Bank

Information Technology Thought Leadership And The Challenges

Christopher Nichols, Director IT/OT Resiliency & Support, Stanley Black & Decker