The Intelligent Warehouse – Automation World

Manufacturing technology advances haveextended the realm of automation beyondplant floor sensors, controllers, visionsystems, and robotics into closely connecteddata collection and analysis using artificial intelligence.These technologies enable a smart factoryto self-optimize and adapt to conditions inreal-time. Despite these advances, warehouse andrelated material moving operations tend not to benearly as modernized as plant floor operations.

If you look inside the most modern environment,warehouse or factory, material handling,broadly speaking, is mostly analog, says MatthewRendall, CEO of Otto Motors, a maker of autonomousmobile robots (AMRs). Any place wherea forklift truck is driving something around, it ishighly analog. That means the amount of data youhave at your fingertips to analyze is limited. Fordecades, operators have been grasping at low accuracy,low frequency, and expensive-to-capturedata trying to figure out how to run a continuousimprovement program.

For example, you can go into a factory or awarehouse today and still see industrial engineerssitting in lawn chairs at an intersection in the plantwith a clip board, pencil, and stopwatch to monitormaterial flows, says Rendall. It is an expensivething to request of a highly trained industrial engineer,so it doesnt get done frequently, he says.

Which means the process, by its nature, is notas exact as it should be; plus, its rarely updated.

But this antiquated, analog surveying methodis shifting in response to the decreasing cost ofcomputer storage, increasing compute power,and new tools that target warehouse and distributionoperations.

The roaming robotOtto Motors is delivering tools to help industrialengineers automate the study of time and motionand it comes in the form of the AMR. Theremarkable contrast that an AMR brings to thetable is that we have sub-inch and sub-secondlevel accuracy, and the marginal cost to collectan additional survey is zero, Rendall explains. Soyou are able to run industrial engineering time andmotion studies on steroids at all times.

The AMRs from Otto Motors are designed formaterial handling in the manufacturing and warehouseoperations in a free-movement manner.In an automotive warehouse, for example, therecould be 30,000 parts moving across the facility.Its a symphony of motion, Rendall says. [Withdata from the AMRs] we are able to harness insightsabout how those 30,000 parts are movingin order to create a smarter traffic network.

The foundation of Otto Motors AMR is a proprietary version of the simultaneous localizationand mapping (SLAM) algorithm, which reliesupon cameras and laser scanners to develop aphotorealistic floor plan.

Think of an AMR as having a photographicmemory. As its driving around it is constructinga picture-perfect representation of that environmentin its memory, Rendall says. Using thatunderlying localization and mapping, you can plotout where a vehicle has traveled. As a result, theside benefit [of an AMR] is that we have one ofthe most sophisticated data collection machines afactory or warehouse has ever seen, patrolling thefloor 24/7, Rendall says.

The AMRs operate with fleet managementsoftware which includes a feature called factoryreplay, providing a time-lapse recap of the entiredays production. The software can take a 16-houroperation and collapse it down to five minutes,which provides an industrial engineer with a birds-eyeview of the floor and the ability to rewind,fast-forward, and zoom in to a certain time duringan incidentas well as before and afterto extractinsights about ultimate root causes of problemson the floor.

Another company, 634AI, offers an AI-driventechnology called Maestro, which is designed tosafely orchestrate the movement of AMRs andcoordinate situational reactions to create a moreefficient and safer environment. Using off -the- shelf, Power-over-Ethernet ceiling-mountedcameras, and proprietary computer vision technology,Maestro pulls video streams from differentcameras to create a grid on the floor.

Those videos are stitched together to create areal-time map of the facility. Then, deep learningAI coupled with computer vision algorithms drawsemantic analytics from data on the factory floor.This data can be productivity-related informationproviding real-time safety alerts, near-missanalyses, task allocation, and the ability to instigateand navigate a robots autonomous capabilitieslive on the factory floor, says Shlomi Hatan,634AIs vice president of business developmentand operations.

Every factor in the process is identified, classified, tracked, and managed by Maestro, includingraw materials, mobile robots, forklifts, boxes, andeven people. Designed to be hardware-agnostic,Maestro is interoperable with other systems toenhance company workflows on a universal scale,Hatan explains. The general rule is, if Maestro seesit, it can be tracked and controlled in real time.

Hatan adds that Maestros control capabilitiescan be extended to deliver custom, AI-generatedproductivity propositions. For example, Maestrocan display forklift slow zones to locate bottlenecksand interferences for vehicles and workers,informing them of obstacles to help shorten traveltimes. It can also track the traveled distance anddriving hours of forklifts and robots to identifyunderutilized resources or assure timely maintenanceof equipment. It can even alert operators ofincorrectly positioned materials and pallets.

Collecting all this information internally andcreating a map of material movement can alsohelp companies with supply chain struggles.Theres an interesting relationship betweentransparency and automation, Rendall says. Wecant influence what time the parts arrive on theloading dock. But if we are responsible for all thematerial handling that happens once the materialshit the receiving dock, we can use a QR code orRFID tag to scan the inventory. That inventory isonly touched by a machine, and when thats thecase, you should be able to see within inch- andsecond-level accuracy where every nut, bolt, andscrew is inside the operation. So you have a muchbetter handle of what inventory you are workingwith inside your operation to more intelligentlyuse the resources available to you.

Automated storageand retrievalIndustrial control suppliers are also adding intelligenceto automated storage and retrieval systems(AS/RS) used in warehouses and distributioncenters.

Beckhoff Automation, for example, offers ashuttle control system for AS/RS applicationsin compact form factors. This system includesTwinCAT Machine Learning software to reduceenergy consumption while optimizing accelerationand deceleration of the shuttles. The machinelearning (ML) functionality automatesthis so there is no human intervention requiredto achieve the process improvements, saysDoug Schuchart, Beckhoffs global materialhandling and intralogistics manager.

In addition, the EtherCAT communicationprotocol used in this system allows informationto be collected in large quantities and stored in adatabase on a Beckhoff controller. That data canthen be transmitted to the cloud.

Once at the enterprise level or the cloud, datascience software can be used to develop ML inferencesfor equipment optimizations and predictive maintenance applications. These ML inferencescan then be deployed using TwinCAT MachineLearning to be executed in the PLC in real time,Schuchart explains.

Dealing with dock delaysAMRs coupled with AI can increase productivityand safety inside operations. And machinelearning coupled with PLCs can improve orderfulfillment and throughput. But what happenswhen the truck doesnt show up at the receivingdock? Traditional warehouse management systems(WMS) dont have the ability to recalculateeverything when there are shipping delays, or ifthere are too many shipments and not enoughdocks, or if inventory is in the wrong building.

To address such issues, AutoScheduler.AI hasdeveloped a cloud-based intelligent warehouseorchestration platform that integrates with existingWMS, ERP, and even yard managementsystems to provide dynamic dock scheduling,proactive cross-docking, and prescriptive analyticsthat balance inventory flow and drive laborefficiencies. Our goal is to be the brain of thewarehouse, says AutoScheduler.AI CEO andco-founder Keith Moore.

The work started as a project with Procter &Gamble (P&G). P&G operates one main plant inOhio that uses seven nearby satellite warehousesfor storage. Across this campus, the projectnoted more than 250 outbound full-vehicleshipments per day and, of those, 85% were dropand hook, and 15% were live load. These operationsrequire manual efforts that are completedbased on who is scheduled and the need for theday. But volatility in the production scheduleand volume made it difficult to plan, resulting inan inefficient operation.

Using the AutoScheduler.AI technology, P&Gdoubled shipments from the plant directly tocustomers without increasing inventory, reducedshuttle moves involving outside warehouses bynearly 50%, reduced workforce planning fromeight hours to 20 minutes per day.

That experience prompted Moore to turn thisinto a scalable system for any warehouse or distributioncenter operation to provide analyticsthat describe what is happening, predict whatwill happen, and prescribe an optimal plan basedon that information.

Every few minutes, AutoScheduler looks at thecurrent situation and then runs what-if scenariosbased on inventory and constraints to maximizeflow through the building. We pull informationfrom the WMS and other systems, run theoptimization, and push it back into the executionsystem. So when someone scans a palletour planflows straight through to the floor, and nobodyeven knows we exist, says Moore.

The AutoScheduler technology is exactly thekind of cognitive toolsets that Stephen Laaper,a principal and manufacturing strategy leader atDeloitte Consulting, says is coming to day-to-daywarehouse operations.

Supply chains will continue to be pressed andstretched in various ways for the foreseeable future,Laaper observes, noting that more actionableinformation is imperative. Because of that,the nature of these solutions are becoming increasinglyimportant.

Read the original post:

The Intelligent Warehouse - Automation World