Chinese Transportation: 25 shippers’ practices slow down the growth of the industry

Chinese Transportation: 25 shippers’ practices slow down the growth of the industry

Lack effective controls over shipments, vendors and ever escalating costs? Tired of handling routine transport complaints?

Do your transport tenders fail to produce expected results?

Unable to differentiate your delivery service and obtain competitive advantage? Take ages to get reliable POD, KPI and freight billing?

Why you cannot get any transport systems to work?

Wonder why trucking industry develops so slowly in China? Read On

These are common issues for shippers and trucking providers in China. They lead to real financial losses for organizations and stressful daily routine for logistics professionals. Low quality and high cost of trucking serv- ices is the usual experience. Conventional wisdom is to lay blame for these on insufficient capabilities of trucking companies. Additionally, attributing them to “this is the Chinese trucking market” factor justifies inac- tion. Indeed traditional trucking & logistics industry develops at glacial speed, in contrast it to break-neck   pace of related sectors of Chinese economy, such as highway construction, express transportation and e- commerce.

We argue here that transport providers (TSP) cannot possibly take full responsibility this situation; neither can they be relied on as sole change agents. Large TSPs are the most visible group in the market, so many natu- rally assume they play a pivotal role. In fact, they are just a thin layer squeezed between clients and vendors and interactions among all these players define market dynamics.

The reality is that trucking always involves at least four parties: shipper, trucking company, driver, and con- signee. In China, particularly within the B2B LTL segment, there are many layers of outsourcing. It therefore is easy to have as many as 8-9 parties involved. This includes at least 3 drivers usually assigned by 3 independ- ent sub-contractors of a TSP.

In this article, we look at typical a B2B shipper’s distribution process, starting from vendor selection through shipment dispatch, track and trace, and ending with performance reporting and billing. Information flow among all these parties involve multiple standards and processes that often will overlap or simply clash with each other, resulting in delays and distortions.

We will contrast it against fast-evolving e-commerce and related express carriers. Conventional understand-  ing of differences between B2B and B2C transport doesn’t provide a clear path for the B2B sector. Shipment size (parcel vs. LTL and FTL) and network ownership are not sole defining factors and B2B shippers and truck- ers can learn useful lessons from B2C area.

Development of road-express companies such as Deppon, CNEX, or TNT-Haou will enable them to capture a bigger market share of the LTL market, in particular for smaller and more profitable shipments. The trend is for shipments to get smaller, due both to evolving consumer behavior and intensive competition. Losing a large and profitable share of business will catch many TSPs off-guard. Many shippers will find themselves imple- menting long overdue data and process enhancements in order to be able to work with road express provid- ers. This will be a rather ironic development because these are factors holding back their current TSP from achieving promised cost and service level improvements.

This article is the first of a series. Over the next few months, we will provide more insight into the actual inner workings of the transport market itself, including the driver’s perspective. We will look into emerging trends in the market. We will publish both long papers like this and shorter articles more frequently. You can help us chose their topics by voting in the survey included at the end of this article. Our content will be available in Chinese and English and will be published on China Domestic Transport Group in LinkedIn, our blog, and our Weixin (WeChat) account. Links to these groups are provided at the end of this article.

Glossary

3PL/LSP – Third Party Logistics Provider/Logistics Service Provider

 B2B – Business to Business – traditional consumer, retail, industrial, and manufacturing companies B2C – Business to Consumer – e-commerce sector

DC – Distribution Centre

ERP – Enterprise Resource Planning System – such as SAP, etc. FTL – Full Truck Load

KPI – Key Performance Indicators LTL – Less Than Full Truck Load POD – Proof of Delivery

RFQ – Request for Quotation – transport tender – vendor sourcing SLA – Service Level Agreement

SOP – Standard Operating Procedure TMS – Transport Management System TSP – Transport Service Provider

WMS – Warehouse Management System

China Worst Traffic Jams Are Not on the Roads – They Are in Your Business Processes.

Manufacturers, brands, and distributors all need to set up suitable physical distribution systems, select qualified transport vendors, process shipments, monitor their execution, pay freight charges, and periodi-  cally review vendors’ performance. The supply chain organization in charge of this process is a support func- tion for internal clients, such as sales and manufacturing. Finance and procurement departments aim to en- sure tthat he lowest cost is obtained through tenders

Such organizational dynamics usually result in adopting nominally high service standards and nominally low freight rates, both of which may impossible to achieve in daily operation. Management intends to stimulate continuous improvements and innovation culture through such dynamics. Unfortunately, in the real corpo- rate world, most people under pressure tend to resort to playing internal KPI games, cut corners, and blame others. Obviously transport vendors will be most convenient scapegoats for internal failings – after all one of the key benefits of outsourcing is that you “outsource” the problems. As a result, these internal failings usu- ally are either ignored or not even noticed.

We believe years of easy double-digit sales growth are gone for most companies selling domestically in China. Therefore, “this is China” hands-off attitude will not cut it for much longer. It is time to take a closer look into the black box of transportation and gradually implement processes that will ensure sustainable competitive advantage.

We have written this article based on our experience in 3PL and logistics software companies over the last 14 years in China. This gave us the opportunity to look into practices of dozens of shippers across many in- dustries. We believe you will be able to identify with many of these issues from your own experience. Youprobably do not see many of them as transport bottlenecks and perhaps consider nothing can be done about them.

Issues below are arranged in 3 phases: planning, execution, and review.

Planning: Process Setup and Trucking Vendor Selection

Delivery coverage, type, and service levels (SLA) need to be set up, and then qualified vendors need to be se- lected and employed. Many daily issues have their root causes here.

Many bottlenecks are avoidable complexities cascaded down to operation and vendors because there is no sufficient thought given to their implications. Sometimes they are even a source of pride and self-validation of  a strict vendor-management system, as if complexity in the transport process was a source of competitive ad- vantage for a shipper. Let’s look at some common examples:

Self-designed, Complex Transport Tariff Formats. Most trucking tariffs in China are based on individual city-to-city lanes. A sliding scale typically will be used with multiple brackets for LTL and FTL shipment sizes. Big shippers will have hundreds, if not thousands, of combinations. Accessorial fees often are added and follow incumbent a TSP structure that may be difficult to follow for new bidders. Often design is   left to people with no real experience and little understanding of implications for operation and ease of billing. Individual shipper tariffs create lot of work for bidders, resulting in low quality bids. Granted, most RFQ are   just benchmarking exercises – but why not benchmark properly? Why not simplify tariffs into zones and get initial quotations for the top 80% volume lanes?

Detailed Shipment Data Not Available or Not Shared. Shippers’ complex tariff format rarely is complemented with comprehensive shipment data useful to new bidders – not just total volume per year/ month, but shipment frequency, size breakdown, and lane allocation. Usually, there is simply no data available from shipper ERP and such data isn’t systematically collected from incumbents. Often the easy excuse of “business confidentiality” is used. The worst case is intentional short-term thinking to get more competitive nominal rates. Whatever the reason, this results in low-quality bids.

Flat Rates with No Minimums. This always appears like a great idea to procurement and finance, especially if most of your shipments are small. TSP will be pushed to “leverage its network” as if lane-haul   was a key cost component in delivering 10kg shipments. In other words, TSP efficiency should result in a cost 100x lower to deliver 10kg than a 1,000kg shipment. As a result, in the real life, TSP or its subcontractors will indeed to “optimize” and batch/delay such shipments and probably fake delivery dates to keep good KPIs.

That is unless an “informal” solution is found to compensate TSP for true delivery cost. Unfortunately, such solutions render RFQ efforts ineffective because bidders usually are not privy to such arrangements.

Late Order Cut-Over Times. Sales pressure to offer high service level dispatch cut-off times are usu- ally set very late. It may even be welcomed by plant or shipper insourced warehouse – why not? DC workers can get much needed overtime by picking orders late. The usual side effect – shipments physically dis-  patched prior to shipment data is transferred – because most DCs in China will rely on manual pick/shipment confirmation often done next morning. As a result, such shipments will anyway spend a night in the vendor’s origin hub. Actual cost increases do not result in any service improvements.

Unrealistic Lead-Times – Sales department feedback about competitor’s standards often determines lead-time requirements. If true, this is usually due to different network set up or order processing flow – it is rarely due to a competitor using faster transport companies. Most TSP will commit to shipper standards to   win business and later work alone or in tandem with consignees or shippers to conceal the reality. If there is  no feedback from consignees, most manual on-time delivery KPI reports will default to 100%.

Dead Standard Operating Procedures (SOP). It is easy for a big organization to get carried away and write lengthy transport SOP, adding unique requirements here and there. It appears like a compre- hensive tool and a sign of solid vendor management practice. It is also much easier than expressing your stan- dards in single-page flowcharts and concise driver instructions on delivery documents. Obviously, after initial writing and training, they just collect the virtual dust on hard drives, are rarely updated, and virtually are never shared with actual, operations level sub-vendors and drivers.

Insufficient Implementation. Transportation is treated as a simple, external activity where TSP should excel on its own. Usually no IT integration is needed. So, unlike in warehouse implementation, it is  brief and often covers just documentation. Even though transport trials are much easier to execute than ware- house it rarely happens. Usually the incumbent TSP will stop or threaten to stop all service immediately after losing RFQ. Most likely he already is working behind the scenes to make the winner fail quickly. So, most is- sues arise during actual rollout and this usually sets the tone for the future relationship between the parties.

Daily Shipment Execution and Monitoring:

Client orders are processed via ERP, WMS, and transferred to a trucking company. TSP will pick goods and deliver them to consignees. Some form of shipment track and trace and Proof of Delivery (POD) usually is pro- vided. More issues will compound the shipper-TSP interface during shipment handover. This is where rubber really hits the road.

Incomplete and Inaccurate Shipment Data. Most shippers in China implement ERP without considering trans- port implications – so, even rudimentary data like full shipping addresses, postal codes, and contact informa- tion are maintained incompletely or simply don’t exist. Gross shipment weight and cbm capacity after packing in a warehouse also is unavailable because many shippers do not even utilize a fully functional WMS module.

As a result, daily order files exported from ERP are incomplete and inaccurate. TSP will be supplied with other “master data” spreadsheets they are supposed to reconcile themselves on the daily basis. Finally, large ship- pers may have multiple ERP sales orders going to the same customer on a given day, and they are rarely con- solidated into shipments out of ERP. TSP has to handle far too much data processing then necessary – result- ing both in longer lead-time and errors that are difficult to trace.

Delayed Shipment Data. It is normal and accept- able in China that shipment data is provided to TSP only dur- ing or even after the physical shipment pickup. The shipper will never be short of excuses why such data is not available in advance – usually “being responsive to client last minute changes” is a solid one. Because there is no IT process gov- ernance, communication is based on Excel spreadsheets  and there is no need to produce such orders in advance for TSP. While this may be a pain for TSP, it certainly looks con-

Shippers often will use trucking to subsidize their in-house warehousing cost. Claiming China’s 2011-14 labor cost environment that it is cheaper manually to load and unload cartons on a vendor truck to increase capacity is simply irresponsible.

venient for shipper. Shippers, however, miss much the larger opportunity of simulating various cost scenarios and using more dynamic allocation methods either to multiple vendors or multiple transport modes. Needless to say, “real-time” shipment pickup notification for consignees also is impossible in such a supply chain.

Loading and Unloading Requirements Even though this is customary practice, a TSP often will find itself baffled by the true scope of “loading” activities that may extend to sorting and virtually picking the goods. In addition, such activity may represent a source of illicit revenue for dock-level staff – a fact somehow most shippers find reluctant to address. Shippers often will use trucking to subsidize their in-house warehous- ing cost. The practice of a trucking company bringing its own loading staff (or worse, hiring freelance labor in the neighbourhood) may be baffling to industry outsiders. There is a clear cost, productivity, and quality advan- tage to perform this activity by your own warehouse staff, in addition to the benefits of controlling dock capac- ity and scheduling better. This practice, however, is so deeply ingrained in the Chinese logistics market that no TSP will question this. It is in particularly “sensitive issue” for “unloading” activities on a client (consignee) side, as we see later.

No Real FTL Loads With a few exceptions (shuttle, milk runs, specialized chemical transport, etc.) there is no such thing as real full-truck loads in China. FTL loads will be more expensive per-unit that corresponding LTL load, even on long distance line-haul. Major reasons are lack of standard truck types, limited palletization, and overloading. Paying TSP by LTL unit enables the shipper to deny any support for or acknowledge over- loading. Real FTL loads also would require shippers to prepare relatively uniform shipments – for example. 30- pallet FTL will not fit 45 pallets (clearly more is not always better) – but with LTL tariff such constraints are han- dled by the TSP. Shippers are solely responsible for not being able to optimize their processes and get true cost KPIs to facilitate FTL moves across plants and DCs. Claiming China’s 2011-14 labor cost environment  that it is cheaper manually to load and unload cartons on a vendor truck to increase capacity is simply irre- sponsible, especially when you consider that most shipments likely will be reloaded many times at transport hubs – wasting time and increasing damage and shortage risk.

Next, let’s look what happens after shipment departure; actual transport seems like deceptively simple activity. It is important to realize that a typical LTL order will pass through the hands of three drivers: pickup, line-haul, and Typical LTL order will pass through the hands of three drivers: pickup, line-haul, and delivery, of which the last one is critical to an overall experience. Most of these drivers are not directly controlled by TSP delivery, of which the last one is most critical to an overall experience. Most of these drivers are not directly controlled by TSP. TSP may use its own pickup driver, but outsource line-haul and final delivery to separate companies. 3PL may outsource the entire delivery to a single subcontractor, but eventually somebody has to separate shipments into legs, outsourced to different vendors. Here is what happens next among different vendors:

Vendor/Driver Management Shippers will have very high expectations on TSP’s ability to “manage”

its vendors, but this term really is an oxymoron in China. Not only vendors and drivers are independent contractors, but their incentives are misaligned. The big problem in China is The big problem in China is not high a percentage of driver-owned trucks but the financial compensation model that pushes the majority of the financial risk down to drivers not high a percentage of driver-owned trucks (also high in Europe and US) but the financial compensation model that pushes the majority of the financial risk down to drivers. TSP rarely assumes pricing risk by paying vendors and drivers for full capacity, for a round trip let alone for monthly km quota. As a result vendors, and drivers are forced to optimize their profit based on factors under their control – such as overloading, delaying/batching shipments, etc. This corner-cutting, in turn, necessitates provides distorted information up the outsourcing chain. Shippers would be better served by replacing the word “management” with “moni- toring” and the TSP may think of incentives rather than penalties for subcontractors in order to move forward.

Outsourcing with Little Monitoring. Unfortunately, effective vendor and shipment monitoring is rarely in place with a TSP. Because of industry compensation models mentioned above, there is little incentive for a TSP to invest in proper controls and certainly there is plenty of resistance from subcontractors. The TSP defaults to the easiest but least effective solution: most track and trace happens via phone and email/XLS  filed by subcontractor which already would be best practice in China. Vendor or drivers don’t need to be pro- active or truthful at all – they are used to being reactive – they have been conditioned by TSPs to act like this. TSPs compensate for their subcontractor “weaknesses” essentially by maintaining their own internal systems (spreadsheets) that are based on verbal feedback. Ironically, this behavior actually enforces current subcon- tractor and driver behavior.

Shippers give little thought how GPS can really be used to track their shipments, especially LTL.

But Most TSP Use GPS! For years, every decent trans- port RFQ in China called for “GPS tracking” and every respect- able TSP has “all trucks equipped with GPS.” China 2012 regu- lations to have GPS installed at registration for trucks over 8 mt capacity lend this statement some truth – irrespective if  they are kept switched on by drivers or not. GPS can be an ef-fective asset and truck monitoring tool, especially when other truck data points are added (engine start/stop, door open/close, temperature, etc.). However, shippers give little thought how GPS can really be used to track their shipments, especially LTL. For this to work, the driver/TSP needs to register, in real time, shipment loading/unloading between trucks and hubs and this information has to be made available to a client in a use- ful, meaningful way.  It is hard to imagine shipper’s staff leaning over computer monitors filled with hundreds   or thousands of shipments tracked on a map on a daily basis – or perhaps several maps provided by different TSPs. Let’s also not forget the simple fact that trucks and GPS units are controlled by various sub-vendors of the TSP, each using different a GPS provider. As a result, GPS tracking is little more than a sales tool. It con- tributes little to tracking information reliability for shipper and nothing at all to the long cycle of POD returns.

Next time ask your TSP how it really uses GPS to keep you informed about your shipments.

But Most Large TSPs Own Their Hubs! More diligent shippers visited their TSP flagship hubs in Shanghai and Beijing. Projection is made that this also applies to “their” 20-50 hubs in other cities that promi- nently feature on network maps that TSPs love to show in their sales presentations. Existence of public de- pots that process perhaps >80% of general LTL cargo on very competitive rates is conveniently forgotten.

Some more honest TSPs (usually ones that lack sufficient client/volume base to make their own hub story plausible) will admit to use of public depots, but always have their “own staff” located there. Of course, such measures are implementable and can be very effective as long as this staff is 1) actually TSP directly hired em- ployees and not usual local subcontractor staff paid and managed by them,  2) well trained in shipper SLA,  and 3) incentivized for truth telling and problem fixing

Lack of Common Language – Shipments have to be handed over between hubs, vendors, and driv- ers using a combination of electronic and paper formats. Electronic information exchange between TSP and vendors involves using various TMSs (optimistic version) or multiple excel files (realistic version). Paper docu- ments will include various TSP,  lane expert, and last-mile delivery company hand-written delivery notes in “fill  a form” standard. Usually none of these have space for shipper requirements so diligently written into ship- per’s SOP and SLAs. Actual delivery dates also may be changed – intentionally or by mistake. Shippers’ ERP delivery note that may serve as POD standard is rarely designed to convey any direct instructions for drivers (often just ERP packing list), so usually it is not much help. Low quality of final delivery thus usually results from low awareness of real shipper expectations, especially because last-mile driver base is dynamic in all but tier 1 cities. On the other hand, these drivers and hub employees have real on-the-ground experience valuable to the shipper, but it is rarely collected and presented to them.

Finally let’s look at shipment delivery. All the previous efforts and issues culminate with final delivery experience for a consignee so driver physical contact will determine the outcome as success, failure, or just another bland experience. Shipper and TSP are miles away with their SOP, SLA and KPIs.

Driver-to-consignee interaction will define if delivery is considered successful. Common issues in- clude:

Driver-to-consignee interaction will define if delivery is considered successful. Low quality of delivery usually results from low awareness of shipper expectations, especially because last-mile driver base is dynamic in all but tier 1 cities

Fragmented Feedback. Consignee usually will call shipper sales/customer service to inquire about delivery status or to complain. Shipper will call its TSP. TSP will call its vendor. Finally, right vendor will call the driver. The process of giving feedback will now repeat in reverse. Can you see how dis- torted, delayed, and effort-consuming the whole process is for TSP and shipper?

Distorted Feedback. It happens that a consignee staff has personal incentive to provide negative feedback to make life difficult to a new TSP. TSPs understand well that consignees’ feedback is an important weapon in retaining old business and incumbents often will resort to this in cases of losing an RFQ. This is easy with no formal and systematic complaint channels and frag- mented information flow; any responsibility for false information can be easily deflected.

Limited Flexibility to react to changing expectations: because of extended and indirect information   flow, consignees will find it difficult to make effective dock arrangements, which costs real money in large DCs or hypermarkets. This results in complaints that could be avoided.

Too Much Flexibility. On the flipside, the, scope of receiving activities, such as extended “unloading,” can be very a effective weapon against a new TSP. Most shippers are passive about it because of difficulties to reconcile the true story. The new vendor will be blamed for the lack of customer understanding and bad service attitude. This costs real money and benefits nobody except the former incumbent TSP hoping to re- capture lost business.

Self-Delayed Deliveries. TSPs do not cause all the delays. Consignees may temporarily lack receiv- ing capacity. In certain industries, it’s a general practice for shipper sales to overload distribution channels to meet their quarterly or monthly quotas. However, it will be convenient to blame the TSP, especially if the ship- per can avoid paying temporary storage or truck detention fees.

Proof of Delivery (POD) – one of biggest issues for shippers in China is to obtain trustworthy docu- ments in an acceptable timeframe. Usually “originals” (paper documents) are required, especially if TSPs were caught faking them in the past. Because it takes a long time (weeks) to collect them from Pan-China deliver- ies, shippers usually ask for “ePOD,” especially for shipments with consignee-reported damage/loss/delay

Such ePOD is usually a form of a fax (not very readable), digital photo, or in the best scenario a high-quality scan made at the TSP hub. Usually it takes days to collect it and if the information on the TSP ePOD is differ- ent from the consignee version (various remarks and dates “mysteriously” tend to appear or disappear there) shipper will have hard time to reconcile this information. It would be simple if the consignee always could be trusted over the driver; but, as we saw before, this is often not the case…

Because most shippers also play the consignee role themselves, their supply chain and transport team could learn many valuable lessons by going down to their own receiving dock, talking to drivers, and seeing their own formal and informal process.

Periodical Reviews

Done usually on a monthly basis, it involves looking at key performance indicators (KPI) to determine opera- tional and financial performance and some improvement plans may be agreed upon. Finally, freight billing is processed between vendor and client so actual fapiao can be issued.

Manual KPI Reports based on Excel spreadsheets usually are prepared by vendors in a one-off, end-of-the- month exercise. Minimum standards will include on-time delivery, damage, and loss rate. For KPI-driven shippers, more data points will be added. Many shippers in China do not invest much effort in KPI design and review. So, it’s common to have TSP KPI 100% because it is common to have 100% stock accuracy in a warehouse managed by an Excel spreadsheet. Even when KPIs are reviewed seriously, spreadsheets are cumbersome to prepare; so, only in very critical performance issues would they be analyzed down to lanes, order types, actual root causes, etc. Few action- able items can be derived from a typical KPI reports used in China.

Manual Billing It may take weeks for new a TSP to assemble billing data in a format demanded by a ship- per, that looked like such a smart idea during RFQ. Process is usually manual – done in a spreadsheet and prone to mistakes – intentional or not. Over months, billing usually can evolve out of the original tariff if the shipper allows adding various ad-hoc and special fees needed to compensate TSP if nominal RFQ rates are too low. Seeds of another useless RFQ exercise will be planted here again. It is amazing that the vast majority of Fortune 500 companies in China run tens and often hundreds of millions of RMB on “free & easy to use”  XLS spreadsheets. A lot of time and energy is focused on nominal cost reductions via RFQ that typically have double if not triple digit percentage variance between bidders and questions are rarely asked about root causes. Instead, it is more convenient to assume that some reputable large local and global TSPs with years  of presence in China still “do not understand the market” and “do not own trucks.” Such shippers shot them- selves in both feet, one representing cost and another service.

It is estimated that for 1 mln spent monthly, vendors facing drivers need to absorb 4-6 mln cash flow. Considering Chinese “shadow-banking” exorbitant interest rates your transport cost is much higher.

Excessively Long Payment Cycle Effective pay- ment cycles usually extend to 90 or even 180 days with many of large shippers. Because total payment cycle for a TSP may include paper POD recovery, preparation, and checking of pro- forma billing plus customary 60-90 days payment terms. Many shippers have informal directives to extend payments as much as possible and the TSP will account for it in its pricing. In fact, TSP subcontractors will account for that, too, in the first place because most TSPs never will pay before monies are received from a shipper. Quite a few TSPs happily will extend vendor payments even more to improve their cash flow and net working capital. Since particular ship- per and TSP “payment reputations” transcend down the subcontractor market, the latter rarely take for granted empty assurances of faster payment. It is estimated that for 1 mln spent monthly, vendors facing driv- ers need to be able to absorb 4-6 mln cash flow. Does it really make financial sense to have small subcontrac- tors fund large shippers? Overall supply chain cost is certainly higher considering Chinese “shadow-banking” extraorbitant loan cost. Sure, shippers and TSPs can do little about nominal payment terms imposed by fi- nance policy. However, plenty can be done by supply chain departments to reduce POD and billing cycle time. Shippers need process and tools exposed in RFQ that would make bidders believe that this time it is really possible to be paid earlier…

High Risk to Change: TSP unsatisfactory cost and operation performance will rarely lead to any changes unless the situation reaches a real crisis level that puts shipper’s staff career at risk. TSP change is a high risk step in China.

Transport arrangements are based on a manual, opaque, and unwritten set of processes and the devil hides in the details –

Changing vendors in China is risky so it is rarely done. As a result development of competitive TSP is slow.

as shown in this article. TSPs serious about a particular RFQ needs extremely good preparation to bid suc- cessfully. Many TSPs find wins rather than victory if the contract is unprofitable and even worse when imple- mentation is a failure with reputational consequences. However, low vendor turnover leads to low industry competitiveness and persistently low service levels experienced by shippers.

Shippers usually expect that one day some “super-qualified” TSP will emerge that will be able to solve issues their incumbents can’t. However, for most of the issues mentioned here, TSP “solutions” canonly be work- arounds leading to the shipper losing control and understanding of their true delivery experience. Such ship- pers end up being overly dependent on the incumbent TSP

This situation is very different in B2C and express delivery sectors. Let’s find out why.

B2C vs. B2B Delivery Model and Experience: B2C Wins Big.

People working for shippers and TSPs can’t help but notice recent success of B2C and related fast-paced de- velopments of Chinese express companies. It is easy to conclude that these transport sectors are completely different for two key reasons: Shipment sizes and unit transport prices are quite different.  Express carriers   can offer standard service levels because their own networks employ drivers directly and run strong internal systems, including IT software and hardware.

We argue that network ownership and employee base is just a liability – a necessary evil – the same as what inventory is to manufacturer or retailer. Likewise, IT hardware and software are expensive to develop and main- tain, but they are a must to enable the process.

Consistent, replicable process is everything. It is a key to profits and customer satisfaction and, therefore, lays the foundation to scalable growth. The key difference with B2B trucking is that  the express process binds clients as much as themselves.

Let’s assume 2 major global industry leaders and fierce direct competitors want to use Shunfeng Express (SF) service in China. SF shipment process, tariffs, and service levels will be considered a given and by no concern will appear be about sharing process or network with direct competitors. A fortune 100 company using an express provider can, of course, get the

The key difference with B2B trucking is that the express process binds clients as much as themselves. Same shipment documentation, same tariff format, standard service levels, same track and trace tools, and best of all same delivery experience. best volume discounts and dedicated key account management, but the underlying process is never custom- ized. They all are based on the same shipment documentation format, same tariff format, same choice of serv- ice levels (express products), same track and trace tools, and best of all, consignees will except exactly the same service level from SF no matter if the delivery is made for competitor 1 or 2.

For this “privilege,” shippers happily will pay SF much higher freight rates than they pay to their B2B truckers, who carry the remaining +95% of their goods. Yes, service levels are higher, but they also demand shipper compliance. It would be unthinkable for a B2B TSP to demand their its process compliance from a shipper in China; they would not even qualify to shortlist. The typical large shipper perception of working with a trucking company is to impose its own standard as validation of doing a good job of strong “vendor management.”    On the other hand, TSPs have little motivation to set such process because they outsource most risks down and so far they usually get away with this.

Consignees, on the other hand, not only expect the same service level from SF no matter who the shipper (supplier), but also they will readily accept seemingly lower service levels. For example, motorcycle delivery is perfectly OK for express shipments but similarly sized shipments should be sent by the closed truck of TSP. Obtaining clean POD can be a hassle for TSPs, but SF delivery will be confirmed without carton contents checking. Any “rude behavior” by a TSP driver immediately will be escalated when nobody would even bother to complain about the staff of a large express vendor. These divergent expectations of shippers and consignees toward express and trucking companies represent an interesting phenomenon and are major limitations for TSPs The situation has some ironic implications for the shippers, too. Eventually, they will be pushed into compliance with processes of express and road- express companies, which will provide more competitive LTL products.

The point is that the asset ownership is necessary to get started. Cloud and mobile technology changes every- thing, and proper application of modern technology is an enabler that would help both shippers and B2B truckers to drive their standards higher. Do you really need an industrial-grade scanner and proprietary in-house system to exe-cute similar process? Is this even possible with multi-tier outscor- ing where drivers deal with orders coming from multiple out- sourcing chains, each with different requirements? Our answer is no in both cases – that is, unless the B2B trucking company de- cides to commit commercial suicide and roll out an express-type asset-based network.

These divergent expectations of shippers and consignees toward express and trucking companies represent an interesting phenomenon and are major limitations for TSPs

So what is holding back Deppon, TNT-Hoau, and CNEX from gaining the mar- ket share from TSP right now?

Obviously these companies initially focus on building scale and replicable process by pursuing small and me- dium shippers, especially because such clients pay higher rates and must settle payments faster. They won’t need to focus their attention on large shippers targeted by TSP, but only until current the SME base offers high growth rates. Time inevitably will come when they turn their atten-

tion to larger shippers and larger shipments offering service and cost better than TSPs. We believe most LTL-oriented TSPs will be caught off guard and by then it will be too late to defend their most profitable chunk of business.

Shippers, on the other hand, may celebrate such developments for as long as they enjoy lower cost and lower internal and external cli- ent pressure because big service provider costs will be considered as a given. Ironically, adopting such standards not only will be bet- ter for shippers as a company, but also for individual supply chain

Supply chain managers will benefit because they will be able to lean on industry standards and fend off often unreasonable requirements and complaints from their sales and manufacturing colleagues. managers who will be able to lean on industry standards and fend off often unreasonable requirements and complaints from their sales and manufacturing colleagues.

However, the path will not be easy considering the low starting point of most shippers in China – most bottle- necks described in this article will have to be eliminated. Most can be eliminated without waiting for market developments and it may be in the best interest of shippers and their incumbent TSP to start right away. Not all shippers would benefit from working with road express companies directly; many still prefer the conven- ience of outsourcing the complete service or geography to a single TSP.

We believe TSP and 3PL, in particular, should build both volume (unique discounts) and management capabil- ity to work with express and road-express vendors and add their capabilities along with other service modes required by their clients, such as FTL or large LTL, etc. This is a successful model in EUR and USA, and such developments will happen in China, probably only faster.

Wider Implications on Industry Development

Now, as we saw a typical B2B trucking delivery process and contrasted this with B2C express/road express, we can look at implications for the trucking industry as a whole.

Sector Consolidation Is Slow and Uncertain.

Investments so far failed to meet investors’ expectations. Spec- tacular failures of U.S. companies focus just on superficial cul- tural reasons – the main point we make here is that economies of scale are difficult to achieve without standard processes that binds clients and independent vendors.  There is a vast supply of subcontractor and driver resources that offer risk-free mar- gins for TSPs. While their quality may be low, there are simply no commercial or operational standards with which to refer.

Therefore, there is no clear payoff to assume management/ ownership controls. There are low barriers for entry into the mar- ket itself and also into a specific contract.

Economies of scale are difficult to achieve without standard processes that bind clients. This also results in low barriers for entry into the market itself and also into a specific contract.

Microsoft Excel spreadsheets remain the most widely used transport management and communication tools in China

Technology Fails Based on our observations, TSP investments in technology are marginal and largely ineffective. Microsoft Excel spreadsheets remain the most widely used transport management and communi- cation tools in China. It seems like a no-brainer- they are free, easy to implement and use, and they are wide- spread even among most basic of companies. However, their very flexibility is one of key reasons of poor serv- ice levels and high costs.

Moreover, traditional transport management software (TMS) focuses on optimization based on non-existing  data and processes and is cost prohibitive to the vast majority of vendors represents very high friction to adop- tion. It adds no value in the current environment effective information flow among all the parties. Using tradi- tional TMS will mean this process still will be fragmented and involve emails, XLS spreadsheets, phone calls, etc. while daily productivity will drop and the system effectively will be abandoned. Therefore, most TMS in China are used just for sales presentations and demos, at best.

Widespread Corruption. It is a no-brainer choice for many – it’s very easy in a 100% manual environ- ment with little effective oversight and opaque outsourcing chains. Separation of procurement activity from the logistics/operations department achieved by most shippers does little in itself to address this. We believe it may even be counterproductive – more sophisticated actions are taken to enable, facilitate, and conceal the

bribery leads to much lower productivity and customer service losses. This – rather than higher direct physical delivery cost – is the real issue here.

It’s also a prevalent business practice of the majority of TSPs. Such TSPs usually overestimate “guanxi fac- tor,” do not invest for the long term, and do not innovate as much as they should. This situation hinders the emergence of large TSPs that could rival road-express carriers in size and quality. Most of large express ven- dors do not need to waste their time and resources finding ways to find “guanxi” and bribe decision makers, which contributes to their faster and more sustainable growth.

Costs Rise Faster than Productivity. Can we expect that productivity gains will offset inevitable escalation of input costs as TSP grows in size?

The mantra of bigger-is-better plays out in industries that provide true economies of scale and replicable stan- dard process that lead to consistent quality. Some logistics sectors such as express or international freight  bear such characteristics and so favor the leaders. Trucking in some developed markets, especially in the  USA, may fit this description better.

Unfortunately, this is not the case in the Chinese trucking industry. Even when the TSP really owns and oper- ates some hubs and lanes, many of the large shipper requirements do not fit there. Such contracts will be op- erated largely out of the TSP core network, on back-to-back contract transfers to subcontractors. This is done to by the TSP to avoid excessive risk and overhead to manage, but comes at a cost of limited leverage. Such TSPs may find little cost and quality synergies from new large clients. It is very common for 3PL and even TSPs in China to have multiple tariffs with the same subcontractor for a different client’s business.

Conclusions

“If you always do what you’ve always done, you’ll always get what you’ve always got”

–   Henry Ford

Outsourcing trucking to large TSP is the only feasible option for B2B shippers. Successful outsourcing relation- ships must be, however, based on a fair allocation of responsibility and effective monitoring (trust but verify).

Shippers who simply “pass the buck” to their TSP and remain hands off cannot seriously expect that such TSPs will behave any differently toward their own subcontractors. This cascading buck passing exercise down to the drivers is the essence of a glacial development speed of improvement of the trucking sector in China.

Trucking in China is and always will be a buyer’s market and therefore clients play a defining role in industry dynamics. Technology and increasing competition among the shippers’ own business environment make trucking a logical target for review and change. Eventually, clients reap what they sow.

We decided to leave it up to readers to consider their own circumstances and experiences first and then draw their own conclusions. Individual company and personal circumstances will vary a lot and it would be too    easy to dispense some generic advice. Feel free to share your opinions (in particular if you disagree with some statements here!) and experience both through the survey included in the article or on our blog. You can sub- scribe to our newsletter or join our LinkedIn group. Note that English and Chinese content usually will not just be a direct translation of each other because authors and audience differ, so you may want to subscribe both.

 

Enhanced Virtual Reality (AR) Technology in Logistics (Part 2)

Enhanced Virtual Reality (AR) Technology in Logistics (Part 2)

3    AUGMENTED REALITY IN LOGISTICS

After the wide range of best practice identified in the four clusters above, we now examine implications of AR in the logistics industry. Although AR is in relatively early stages of adoption in logistics, it could offer significant benefits. For example, AR can give logistics providers quick access to anticipatory information anytime and anywhere. This is vital for the prospective and exact planning and operation of tasks such as delivery and load optimization, and is critical to providing higher levels of customer service.

At DHL Trend Research, we are transferring to logistics what we see as best practice in other industries, and we are envisioning several use cases for AR in the logistics industry. These serve as a visionary outlook and the basis for discussion rather than a concrete prediction of how AR in logistics will develop in the future.

These use cases are arranged in the following categories:

  • Warehousing Operations
  • Transportation Optimization
  • Last-mile Delivery
  • Enhanced Value-added Services

3.1        Warehousing Operations

AR has so far shown most promise for logistics in warehousing operations. These operations are estimated to account for about 20 % of all logistics costs, and the task of picking accounts for 55 % to 65 % of the total cost of warehousing operations.6 This indicates that AR has the potential to significantly reduce cost by improving the picking process. It can also help with the training

of new and temporary warehouse staff, and with warehouse planning.

Pick-by-Vision: Optimized Picking

 

  • Picking staff are equipped with wearable AR devices for the picking process
  • The solution offers digital navigation to find the right route and item more efficiently, while reducing training time
  • Main objectives: reduce picking errors and search time

In logistics, the most tangible AR solutions are systems to optimize the picking process. The vast majority of warehouses in the developed world still use the pick-by- paper approach. But any paper-based approach is slow and error prone. Furthermore, picking work is often undertaken by temporary workers who usually require cost-intensive training to ensure they pick efficiently and without making errors.

Systems by Knapp, SAP, and Ubimax are currently in the late field-test phase and consist of mobile AR systems such as a head-mounted display (HMD), cameras, a wearable PC, and battery packs that provide enough energy for at least one work shift. The vision picking software offers real-time object recognition, barcode reading, indoor navigation, and seamless integration of information with the Warehouse Management System (WMS). A key benefit of vision picking is its provision of hands-free intuitive digital support to workers during manual picking operations.

By using a system like this, each worker can see the digital picking list in their field of vision and – thanks to indoor navigation capabilities – see the best route, reducing their travel time by efficient path planning. Using automated barcode scanning capabilities, the system’s image recogni- tion software (e.g., provided by Knapp KiSoft Vision7) can check whether the worker has arrived at the right location, and guide the worker to quickly locate the right item on the shelf.

The worker can then scan the item and register this pro- cess simultaneously in the WMS, enabling real-time stock updates. In addition, such systems can reduce the amount of time required to orientate and train new employees, as well as bridge any language barriers with migrant workers.

Field tests of these AR systems have proved they offer significant productivity improvements in warehousing operations. For example, constant picking validation can decrease errors by as much as 40 %. Although today’s picking error rate is very low, even using a pick-by-paper approach – experts estimate a rate of 0.35 % – every error must be prevented, because it typically results in high follow-up costs.8

Warehouse Planning

 

  • Creating a mixed-reality simulation of warehouse operation processes
  • Modifications are overlaid in the real environment to ‘field test’ and adjust planned redesign measures
  • Main objective: support and reduce the cost of warehouse redesign and planning

AR is also likely to affect warehouse-planning processes. Today’s warehouses are not only used as storage and distribution hubs; more and more, they house a growing number of value-added services, ranging from product assembly to product labelling, repacking, and repair.

This means hubs must be redesigned to accommodate these new services. AR can be used to visualize any planned rearrangements in full scale, making it possible to place interactive digital representations of proposed future modifications in the present, real warehouse environment. Planners can test whether measurements of a planned modification will fit in place, and model new workflows. In future, this could allow a real warehouse to be used as the test bed for warehouse operation planning.

3.2    Transportation Optimization

Over the last decade, the use of advanced information technologies by logistics providers has greatly improved the efficiency, reliability, and security of freight transport- ation. AR has the potential to further optimize freight transportation in areas such as completeness checks, international trade, driver navigation, and freight loading.

Completeness Checks

  • AR devices register if a delivery is complete and ready for pick-up
  • Capturing pallet and parcel numbers and volume using markers or advanced object recognition technology
  • Automated confirmation of pick-up by AR after the correct number of undamaged parcels is recognised
  • Main objectives: time savings, completeness check, damage detection

AR can achieve more effective pick-ups. An AR-equip- ped collector could quickly glance at the load to check if it is complete. Currently, this requires manual counting or time-consuming barcode scanning with a handheld device. In the future, a wearable AR device could use a combination of scanners and 3D depth sensors to deter- mine the number of pallets or single parcels (by scanning specific markers on each parcel) or their volume (using measurement devices). This measurement is compared to

predefined values and the result – hopefully a match – will be displayed to the collector. This AR system could also scan items to detect any damage or faults.

International Trade

  • AR support for global trade service providers
  • AR devices can check (printed) trade documents and identify commodity code classification
  • Real-time translation of parcel labels or foreign trade terms
  • Main objectives: facilitate trade documentation and international freight handling

With more of the world’s regions poised to flourish economically, transport flows to and from emerging markets are increasing significantly. This represents a large opportunity for logistics providers but it also increases complexity, as there is significant variation in trade regulations and requirements around the world.

AR is likely to prove valuable for providers of global trade services. Before a shipment, an AR system could assist in ensuring the shipment complies with the relevant import and export regulations, or trade documentation has been correctly completed. An AR device can scan trade docu- ments or goods for key words and automatically propose changes or correct the commodity code classification.

After shipment, AR technology can significantly reduce port and storage delays by translating trade document text such as trade terms in real time (see the Word Lens app in section 2.1).

Dynamic Traffic Support

  • Replacement of navigation systems in delivery vehicles with AR devices (glasses or windshield projection)
  • Analysis of real-time traffic data and display of relevant information (e.g. blocked or alternative routes) in the driver´s field of vision
  • Superimposing critical information on surrounding, vehicle and cargo (e.g. temperature of cold store)
  • Main objective: optimized routing on the fly, improvement of driving safety, minimizing of driver distraction

Traffic congestion often prevents the smooth running of many economic processes that heavily depend on the smooth flow of physical goods. It’s estimated that traffic congestion costs Europe about 1 % of gross domestic product (GDP) each year.9 And as congestion increases, there is high demand for solutions to improve punctuality.

In future, we will see increasing use of dynamic traffic support with real-time traffic data to optimize routes or re-route shipments on the fly. AR driver assistance apps (either with glasses or a windshield display) could be used to display information in real time in the driver´s field of vision. In effect, AR systems will be the successors to today’s navigation systems, with a key advantage that the driver doesn’t have to take their eyes off the road. AR systems can also provide the driver with critical information displays on their vehicle and cargo (e.g, confirmation of cargo temperature).

Freight Loading

 

  • Use of AR devices for optimized cargo loading
  • Loader receives load plan and instructions (which pallet to take next and where to put it) directly on their AR device display
  • Renders printed load lists unnecessary
  • Main objective: speed up the freight loading process

Today, freight transportation by air, water, and road makes extensive use of digital data and planning software for optimized load planning and vehicle utilization. Issues such as content, weight, size, destination, and further pro- cessing are taken into account for each item. Even though there may be some potential for further improvement, the bottleneck is often the loading process itself.

AR devices could help by replacing the need for printed cargo lists and load instructions. At a transfer station, for example, the loader could obtain real-time information on their AR device about which pallet to take next and where exactly to place this pallet in the vehicle. The AR device could display loading instructions, with arrows or high- lights identifying suitable target areas inside the vehicle.

This information could be generated either in advance by planning software or on the spot by ad-hoc object recognition. The latter approach is comparable with the popular computing game Tetris, where the gamer has to place the next random item according to its shape in order to maximize available space and avoid gaps. In contrast to current paper-based lists, AR-supported cargo lists would also allow for real-time – something that happens quite often during the loading process.

3.3        Last-mile Delivery

Another important field of application for AR is at the last mile.10 The growing use of e-commerce has led to a boom of last-mile delivery services, which is the final step in the supply chain and often the most expensive one. Therefore, the optimization of last-mile delivery to drive down product cost and increase profit is a promising field of application for AR devices.

Parcel Loading and Drop-off

Estimates suggest that drivers spend between 40 % and 60% of their time away from the distribution center not driving. Instead, they spend much of this time locating the correct boxes within their truck for the next delivery. Currently, to find a box, drivers must rely on their memory of the loading process.

In future at the distribution center, each driver could receive critical information about a specific parcel by looking at it with their AR device. This information could include the type of goods being transported, each parcel’s weight, delivery address, and whether it is fragile or requires specific positioning to avoid damage. The device could then calculate the space requirements for each parcel in real time, scan for a suitable empty space in the vehicle, and then indicate where the parcel should be placed, taking into account the planned route.

With efficient and intelligent loading, and with AR devices highlighting the right parcel for the driver, the search process would be much more convenient and significantly accelerated at every drop-off.

In addition, AR could help to reduce the incidence of package damage. One of the key reasons why parcels get damaged today is that drivers need a spare hand to close their vehicle door, forcing them to put parcels on the ground or clamp them under their arm. With an AR device, the vehicle door could be closed ‘hands-free’ – the driver could give a voice instruction or make an eye or head movement (see the SixthSense and Revolv apps in section 2.3).

Parcel Loading & Drop-off

  • Equip staff with wearable AR devices for parcel handling, loading, and delivery processes
  • Through AR, all parcels are overlaid with critical information (e.g., contents, weight, and destination) and handling instructions, and parcels are loaded intelligently into the vehicle
  • Main objectives: improve parcel handling, avoid improper handling, ensure load optimization

Last-meter Navigation 

With the vehicle door shut and the correct parcel in the driver’s hands, often the next challenge is to find a specific building. This is particularly true when doing a first-time delivery, as there can be many complicating factors such as obscured or missing house numbers or street names, entrances hidden in backyards or, as is the case in many developing countries, no structured naming scheme for streets and buildings.

AR could be extremely helpful here; the driver could point an AR device at a building or block of buildings and it could display information such as a Google Street View or relevant details from other databases. When there’s no available public database, particularly with information on the position of entrances or other local features, the AR device could also be used to place markers, thereby building up an independent database over time. At the next delivery to the same address, the AR device could access this previously collected data; virtual layers of information could be created accordingly.

Sometimes, last-meter delivery requires indoor navi- gation. While GPS-based navigation works well out in the open, buildings sometimes cause severe interference to the GPS signal. A solution could be for the learning AR device to place LLA (latitude, longitude, altitude) markers at various internal points.

  • AR-supported identification of buildings and entrances, as well as indoor navigation for faster delivery
  • A learning system that is able to add user- generated content, particularly when public databases are unavailable
  • Main objectives: efficient indoor navigation, reduce search and delivery time, especially for first-time deliveries

AR-secured Delivery

 

  • AR-based unambiguous identification of the parcel receiver using face-recognition technology
  • Visual approval/refusal instead of ID card or signature
  • Main objectives: improve security of registered letters, speed up the delivery process
  • Service would require approval and registration in advance

Equipping staff with AR devices could also increase security and improve the quality of customer contact. Using facial-recognition technology, the person receiving a parcel could be unambiguously identified without having to show any ID. The AR device could take a picture and automatically match this in a secure database. Due to data privacy issues, it would be necessary for the recipient to give prior permission for use of this AR face-confirmation technique. This service may not be applicable for ordinary every-day deliveries, but when the parcel has extraordinary high value, users may appreciate this enhanced level of security as it is superior to an easily forged ID card or recipient signature.

3.4    Enhanced Value-added Services

As well as helping logistics providers to improve their processes, AR can also enable them to perform new services for their customers, such as assembly and repair, and provide new customer support tools.

Assembly and Repair

  • Assembly and repair teams are equipped with hands-free AR devices (glasses) and software that support specific tasks
  • The software blends in visual step-by-step instructions for the assembly or repair while keeping each worker’s hands free to conduct these steps
  • Main objectives: control quality, significantly reduce training costs

 

More and more logistics providers offer added value to customers with services such as assembly and repair. For example, DHL not only collects materials from component providers for Audi, but also assembles these components into interior door panels that are

then delivered to the Audi production plant in Germany.

Currently, skilled workers are required for such tasks, and each must be individually trained. However,

in future AR could train and aid warehouse staff to assemble a variety of products and ensure that high standards of service are maintained, potentially reducing cost for customers.

The AR system could ensure quality control by monitor- ing each work step (via enhanced image recognition) and detecting errors in the assembly process. For repair staff, it could offer an intuitive and visual way to support the identification and fixing of errors, especially with the ever-increasing number of end-consumer technologies and gadgets. The use of such interactive repair guides could significantly reduce training costs as well as the technical staff ’s average repair time.

Customer Services

In the near future, AR-enhanced parcel service applica- tions could enable customers with an AR-capable device to volume scan the measurement of goods to be shipped and estimate the weight to establish the perfect size and lowest price parcel box from their logistics provider. In addition, this app could display different shipping and insurance price options.

While such an elaborated app is not yet available today, there is a simpler version in use. The DHL Paketassistent11 lets the user print a sheet containing an icon that is similar to a QR code. Using a webcam, holograms of available DHL parcel boxes are projected for customers to then match their items to the right-size box.

In conclusion, AR has a promising future in the logistics industry. Ranging  from  picking-by-vision in warehouses to assisting customers with after-sales activities, it is clear that AR can play a part in almost every step of the logistics value chain. Although only a few of these use cases are currently being developed,

there are encouraging first signs of AR adoption in the logistics industry. This trend will continue to grow, and we hope that more logistics providers will participate to drive the AR revolution.

  • AR apps for end-consumer devices such as smartphones and tablets for a convenient shipping experience
  • Main objectives: help customers determine and order the correct shipping options by scanning the goods to be sent and overlaying this scan with a virtual representation of shipping boxes; improve parcel handling

OUTLOOK

It may be difficult to imagine smart glasses becoming an essential part of our daily appearance. And, as is often the case with new and emerging technologies, it isn’t easy to gauge whether there will be a rapid uptake of AR. But it is quite possible that AR devices will one day be as normal and widespread as smartphones.

We hope this trend report has helped to illustrate that AR is no longer in the realm of science fiction. As demonstrated by our wide variety of best practice use cases, AR is already providing tangible benefits across many industries today, including logistics.

But before AR devices (especially wearable ones) can be widely adopted in logistics, we need to overcome a number of technical and societal challenges including battery life, high investment cost, network performance

issues, privacy, and public acceptance, to name just a few.

Nevertheless, logistics providers and their customers should be aware of the benefits AR can offer now and in the future. We must be ready to take advantage of opportunities as they arise, many of which are currently untapped.

Looking ahead, AR is well positioned to deliver some of the future´s most intriguing user interfaces and display technologies, harnessing the potential to fundamentally change how we perceive information and interact in

our professional and private lives.

We are clearly in the early stages of what is sure to be an exciting journey that could result in the integration of AR into daily life in logistics – so come join us and together let’s look at reality in new ways.

 

Phần 1: Giới thiệu về AR
—–
Nguồn: DHL report

Technology logistics (Part 1)

Technology logistics (Part 1)

1     UNDERSTANDING AUGMENTED REALITY

Imagine your car breaks down in the middle of the highway. You know very little about vehicle mechanics, and the next garage is miles away. Today, this breakdown is likely to cost you lots of time and money to fix; but tomorrow, it may be no more than a minor glitch in your day. Using your smart glasses, you’ll be able to launch your repair app and assess the problem through your normal line of sight accompanied by a step-by-step repair guide for your particular make and model of car, without the help of a mechanic.

Or imagine you’re out shopping, and want to know how other consumers have rated a jacket that you’re thinking of buying. By glancing down at the product with your smart glasses, you’ll instantly see extra information displayed alongside the jacket – user ratings, product price range, and supply information – all of which empowers your purchasing decision.

This is Augmented Reality (AR) – where every object you see could be enriched with additional and valuable information. AR is defined as the expansion of physical reality by adding layers of computer-generated informa- tion to the real environment.1 Information in this context could be any kind of virtual object or content, including text, graphics, video, sound, haptic feedback, GPS data, and even smell. But AR is more than a simple displaying technology. It also represents a new type of real-time natural user interface for human interaction with objects and digital devices.

AR is made possible by performing four basic and distinct tasks, and combining the output in a useful way.

Figure 1: Basic Functionality of Augmented Reality; Source:

  1. Scene capture: First, the reality that should be augmented is captured using either a video-capture device such as a camera, or a see-through device such as a head-mounted
  2. Scene identification: Secondly, the captured reality must be scanned to define the exact position where the virtual content should be embedded. This position could be identified either by markers (visual tags) or by tracking technologies such as GPS, sensors, infrared, or
  3. Scene processing: As the scene becomes clearly recognized and identified, the corresponding virtual content is requested, typically from the Internet or from any kind of
  4. Scene visualization: Finally, the AR system produces a mixed image of the real space as well as the virtual content.

Experts also differentiate between Augmented Reality and Virtual Reality (VR). VR is a completely computer- generated, immersive and three-dimensional environment that is displayed either on a computer screen or through special stereoscopic displays, such as the Oculus Rift.

In

 

1.1        From Digital Gimmickry to Revolutionizing Business?

Despite the surge in widespread media coverage over the past 12 months, the majority of AR solutions that we read about today are still in development. Only a few hardware solutions are being mass produced and readily available to purchase off the shelf.

Just a couple of years ago, there were only a handful of available commercial AR applications – in fact the first AR app for the iPhone was not released until 2009.2. In 2011, global AR revenues were as low as USD 181 million3 and, at that time, AR was often perceived by the public as just a marketing gimmick: a technology in search of a useful application. There was little public awareness, and applications were primarily developed to gain quick PR wins, or their value was limited to attention-grabbers such as adding video effects.

However, latest forecasts predict that by 2017 the AR market will grow to USD 5.2 billion – an impressive annual increase of almost 100 %. With substantial funding being poured into AR projects and start-ups, especially by large corporations such as Google, Canon, and Qualcomm, we can expect the first significant wave of consumer-ready AR products to be launched over the next 12 months. And with concrete business benefits coming to light, experts are convinced that AR will be the ‘next big thing’ in the consumer, medical, mobile, automotive, and manufacturing markets.

AR is no longer just a marketing ploy. We will see continued uptake of AR and, as it grows, its application will be accelerated by technological progress.

Figure 2: Global AR Revenues 2012–2017 (estimate); Source: Xcubelab

1.2        Hardware Overview

 Both the development and implementation of AR software solutions rely, in turn, on the development of suitable and robust AR hardware platforms. And this hardware development is driven by technological progress in the fields of computer processors, displays, sensors, mobile Internet speed, battery life, and more. By looking at the types of AR platform currently available, and predicting what lies ahead, the following AR items can be identified today:

  • Handheld Devices
  • Stationary AR Systems
  • Spatial Augmented Reality (SAR) Systems
  • Head-mounted Displays (HMDs)
  • Smart Glasses
  • Smart Lenses

Handheld Devices

Figure 3: Smartphone Example of a Hand- held Device; Source: Freshmindstalent

We are currently experiencing a massive boom in Handheld Devices such as smartphones and tablet computers, and this will accelerate AR adoption. These devices are appearing with ever-better features such as higher-resolution displays, more powerful processors, and high-quality cameras, along with a growing array of sensors providing accelerometer, GPS, and compass capabilities, making them very suitable AR platforms. Although handheld devices are the easiest way for consumers to access AR apps, most are not wearable and so they cannot give users a hands-free AR experience.

Stationary AR Systems

Figure 4: Stationary AR Wardrobe at a Topshop in Russia; Source: Mashable

Stationary AR Systems are suitable when a larger display or higher resolution is required in a permanent location. Unlike mobile AR devices, these motionless systems can be equipped with more advanced camera systems and can therefore provide more precise recognition of people and scenes. Moreover, the display unit often shows more realistic pictures and is not so much affected by environmental influences such as sunlight or dim lighting.

Spatial Augmented Reality (SAR) Systems

Figure  5:  SAR  System  at  Volkswagen;  Source:    Volkswagen;

In contrast to all other systems, Spatial Augmented Reality (SAR) Systems include virtual content directly projected on top of the real-world image. SAR systems are frequently stationary in nature. Any physical surface such as walls, desks, foam, wooden blocks, or even the human body can be turned into an interactive display.

With projectors decreasing in size, cost, and power consumption, and with progress in 3D projection, what’s emerging is a completely new range of interaction and display possibilities. The biggest advantage of SAR systems is that they provide a more accurate reflection of reality, as virtual information can be visualized with actual proportions and size. Furthermore, content can be made visible to a larger  number of viewers, and this can for example enable simultaneous working.

Head-mounted Displays (HMDs)

Figure 6: Canon`s Mixed Reality Headset as an HMD Example; Source: Digitaltrends

Head-mounted Displays (HMDs) represent another rapidly growing AR hardware item. HMDs consist of a headset, such as a helmet, which is paired with one or more (micro-) displays. HMDs place images of both the physical world and virtual objects over the user‘s field of view. In other words, the user does not see reality directly, but sees an (augmented) video image of it. If the display is placed only in front of

one of the user´s eyes, it is called a monocular HMD (in contrast to binocular systems, where both eyes view the display). Modern HMDs are often capable of employing sensors for six degrees of freedom (allowing the user to move their head freely forward/backward, up/down, left/right, pitch, yaw, and roll). This enables the system to align virtual information to the physical world, and to adjust according to the user‘s head movements.

Smart Glasses

Figure 7: Vuzix M100 Smart Glasses; Source: Vuzix

Many companies from the consumer electronics industry are expecting Smart Glasses to be the next global consumer hit after smartphones. These AR devices are in essence glasses equipped with screens, cameras, and microphones. With this

concept, the user’s real world view is intercepted and an augmented view re-displayed in the user’s field of vision. AR imagery is projected through or reflected off the surface of the eyewear lens pieces. The most prominent examples of this technology are Google Glass and Vuzix M100. However, one of the most exciting smart glasses developments today is the Atheer One – these smart glasses are equipped with

3D depth sensors, allowing users to physically control the virtual content displayed in front of them

 

Smart Lenses

Figure 8: Lens Containing Metal Circuit Structures, Developed at the University of Washington; Source: Washington

Glasses are certainly not the end of the story. Research is gaining momentum into Smart Lenses that can display AR imaging; companies such as Microsoft and Google are busy unveiling their own smart lens projects.

The idea is to turn conventional lenses into a functional system by integrating control circuits, communication circuits, miniature antennas, LEDs, and other optoelectronic components. In future, hundreds of integrated LEDs could be used to form an image directly in front of the eye, transforming the lens into a display. However, before this can become reality, a couple of significant challenges must be overcome, such as how to power the lenses, and how to ensure that the human eye is not damaged

2   AUGMENTED REALITY BEST PRACTICE

In this section, we explore how this emerging techno- logy is currently being used across different sectors, and anticipate best practice that’s likely to become mainstream in future. We have selected a number of innovative AR examples, clustered into four functional categories; each provides individuals or companies with significant benefits when using AR applications.

 

2.1  Context-sensitive Information – Information at the Right Time and Place

The first cluster is context-sensitive information, encompassing various applications that enable easy access in context-specific situations to static data that’s already available on the Internet.

Figure 9: Wikitude; Source: Wikitude

Wikitude and Metaio’s Junaio are two leading examples  of AR browsers that provide context-sensitive information software capable of recognizing locations or objects to link digital information to real-world situations. The software runs on any smartphone and displays additional digital information about the user’s surroundings in a mobile camera view.
This additional digital information could be nearby places of interest, for example, such as museums, shops, restaurants, or the pedestrian route to the next bus stop. The software includes image recognition, user position localization via GPS and WiFi, and 3D modelling.

Figure 10: Word Lens; Source: Questvisual

One of the most promising areas of application in AR is the field of language translation. An existing app is Word Lens software which runs on almost any smart phone and simultaneously translates text from one language to another. With this app running, the user merely points their device to a piece of text written in a foreign language. Then their device displays this information translated into the user’s native language. It is written in the same font and on the same real- world background as the original text.

Figure 11: Infinity Augmented Reality App; Source: Infinity

Another example of easy access to Internet information in context-specific real-life situations is the combination of face detection and AR. An app that is promised to be available soon is the Infinity AR application. The concept is to analyze a face and compare and match it to profile pictures found in social networks (e.g., Facebook). Infor- mation posted in the matched profile is then displayedin the user’s field of vision.

As well as being useful in consumer applications, this technology is very promising for law enforcement agencies (e.g., scanning crowds for wanted criminals). Under- standably, this application has also raised many privacy concerns.


Figure 12:
Volkswagen MARTA; Source: MARTA

A highly practical solution to the best practice of provi- ding the right information at the right place in the auto- motive sector is the MARTA (Mobile Augmented Reality Technical Assistance) system developed by VW. This system comes in handy when a car isn’t running properly, helping its user to perform vehicle repairs and maintenance. It recognizes vehicle parts via object recognition, and describes and pictures all required repair and maintenance steps in detail and real time, along with information about any equipment requirements. This app runs on various mobile devices. Currently, the system is for the exclusive use of VW Service, but it is conceivable such systems could become available for consumer mar- kets in future, helping everyone to fix their cars without knowing very much about mechanics.

2.2    Enhanced Senses – Becoming Human 2.0

Even today, AR applications can offer much more than just retrieving Internet information on the go. The following AR use cases enhance reality using newly generated information from data gathered mostly by the device’s sensors. They feature a range of devices that enhance the senses, extending human capabilities beyond our current achievements.


Figure 13:
Recon Jet; Source: Recon Instruments

Recon Jet is an already available AR system for leisure activities. The device’s sports-oriented heads-up display connects to third-party sensors such as Bluetooth and WiFi, and offers navigation and weather information, access to social networks, and real-time information about performance – for example, a runner would want to know their speed, distance to the finishing line, current elevation gain, and their heart rate. With these capabilities, the Recon Jet points towards the future development of wearable AR that can monitor the vital signs and surroundings of people working in hazardous environ- ments or physically demanding jobs.


Figure 14:
BMW HUD; Source: BMW

Another heads-up display (HUD) is used to project sensory information such as driving speed onto the windshield of some BMW cars. This enhanced-senses capability has been used by the automotive company since 2004, and BMW is constantly working to improve this HUD with additional features. BMW’s current ConnectedDrive HUD is augmented by virtual markings that are superimposed on real objects in the external environment. This allows navigation infor- mation or information from driver assistance systems to be displayed in exactly the right position on the driver’s view of the road scene. Navigation instructions can be blended into the road, and vehicles or safety-relevant objects can be highlighted or marked in context. A great example is the visual information provided by BMW’s night vision system.


Figure 15:
iOnRoad; Source: iOnRoad

A similar but less advanced augmented driving assistance system for the mass market is the award-winning iOnRoad app. Using only the smartphone camera and some vision algorithms, it provides real-time features such as collision warning, headway monitoring, off-road warning, and a black box video recording function which can come in handy after an accident.


Figure 16:
Liver Explorer; Source: Fraunhofer

In a completely different field of application, surgeons can access enhanced senses with the Liver  Explorer app by the developer Fraunhofer MEVIS. This app provides real-time AR guides and assistance to the medical prac- titioner. The device’s camera films the liver and, using AR, superimposes surgical planning data onto the organ. In addition, the software can react in real time (e.g., updating the surgical plan according to the movement of blood vessels which the system tracks constantly). These capabilities go beyond the MARTA system’s provi- sion of context-sensitive information. Assuming the app receives positive evaluation, it is likely to be modified for future expansion into additional surgical fields.


Figure 17:
Q-Warrior Helmet; Source: Telegraph

In dangerous situations it is especially important to have crucial information at hand. Therefore the military is one of the biggest investors in AR applications. One military app is the Q-Warrior  Helmet. This AR item is intended to provide soldiers with “heads-up, eyes-wide, finger-on- the-trigger” situational awareness, friend-or-foe identifi- cation, night vision, and an enhanced ability to remotely coordinate small units. The helmet transmits detailed positional information about each wearer to the others, allowing the system to gather, map, and share information and positions in real time on the battlefield and during reconnaissance. It is easy to look ahead and anticipate similar systems being developed for other professionals working in dangerous environments, such as fire fighters and law enforcement personnel.

 

2.3       Mixed-reality Simulations – Exploring the Virtual in the Real

 While the above examples augment reality by providing static digital information, this next AR cluster goes a step further. These so-called mixed-reality simulations allow users to dynamically adapt or change virtual objects in the real environment. Uniqlo’s Magic Mirror offers an even-more personal AR fitting experience. Introduced in 2012 in a Uniqlo shop in San Francisco, USA, this large augmented mirror recog- nizes the shopper’s size and selected fashion item, so there’s no need to try on different colors. The shopper simply puts on one item and steps in front of the mirror; a touchscreen then prompts the consumer to select other available hues, and projects back the modified reflection.

 

 

Figure 18: IKEA AR App; Source: IKEA

One of the most prominent examples is the latest Ikea catalog. Developed by Metaio, this AR app lets con- sumers use their mobile devices to ‘place’ digital versions of selected Ikea furniture in their real living rooms, making it easy to test whether the dimensions, style, and color of the furniture fit in a chosen position. This app also allows the user to change the size and color of each piece.


Figure 19:
The Magic Mirror; Source: Trendhunter

Figure 20: MREAL; Source: Engadget

The Mixed Reality System (MREAL) by Canon supports the design process by enabling the seamless merging of 3D computer-generated models with real- world objects in a real environment. For example,

it can help with designing a new model of car in the automotive sector. MREAL allows multiple users to work collaboratively and simultaneously on a full-scale product design. The system can be used to analyze how real components will fit together with a newly planned design. It does this by creating a 3D model of both the existing components and the new concept, and then brings both together.

For example, an existing car seat can be integrated into the projection of a virtual new car design. Since MREAL delivers mixed reality, users can actually sit in the (real) seat and see both the real environment outside the car along with the digital representation of the car interior, including the planned new dashboard and steering wheel.

 

Figure 21: MiRA; Source: Highflyer

Another industrial AR app that’s already in use comes from Airbus. With the master for a new aircraft production process developed entirely with digital tools, Airbus colla- borated to create the MiRA (Mixed Reality Application) in 2009. This app increases productivity in production lines by using AR to scan parts and detect errors. On the A380, MiRA, which today consists of a tablet PC and a specifically developed sensor pack and software, has reduced the time needed to check tens of thousands of brackets in the fuse- lage from 300 hours to an astonishing 60 hours. Further- more, late discoveries of damaged, wrongly positioned or missing brackets have been reduced by 40%.5


Figure 22:
The Interactive Hatsune Miku; Source: Hatsune Miku

Our final example in this AR cluster gives a glimpse of what we can expect of AR apps in the mid-range future. A hacker from Japan used an available 3D model and cheap motion sensors to have an AR ‘date’ with the famous virtual Japanese pop star Hatsune Miku. In his video, he shows how he ‘walks’ with Miku in a real park and how Miku recognizes and reacts to real-world objects (e.g., by sitting on a real chair). This software even makes it possible to interact with the virtual pop star (e.g., touching her tie or head). While this application is clearly sensationalist it is more than just a gimmick. It gives the idea that soon people may be accompanied by virtual companions who could provide assistance when needed (e.g., in medical or engineering tasks, or as a human-like interface for everyday digital issues such as managing a personal calendar, notes, and contacts).

2.4    Virtual Interfaces – Controlling the Real Through the Virtual

With more and more ‘smart’ objects connected to the Internet, and with new ways of accessing digital informa- tion, more and more people want to work with AR devices and data. Therefore, our fourth cluster, virtual interfaces, focuses on AR technologies that offer new options for controlling real-world objects through digital means. Essentially, this allows a mixed reality where real objects can be altered and controlled.

Figure 23: SixthSense; Source: Pranavmistry

An advanced way to interact with the digital world on the go is to use gestures. One example of a gestural inter- face system is SixthSense, developed by MIT. While this system currently uses spatial AR technology, it can also be used with all other technologies. The system allows the user to interact with information via natural hand gestures. In order to capture the intended input of the user, the camera recognizes and tracks the user‘s hand gestures using computer-vision-based techniques.

 

Figure 24: Revolv; Source: Revolv

AR-based interfaces are not limited to computer devices. They can be used to control cars, entertainment, and household appliances such as heating systems. One example is the home automation system Revolv, which is still under development. In combination with Google Glass, the system gives the user control over all digital devices in the household (e.g., the light system and locking system). The result is an augmented ‘smart’ household environment, which can be remotely controlled by voice or fingertip.

Virtual interfaces can go beyond the home, as shown by Yihaodian, the largest food e-retailer in China. The company recently announced that it was going to open up the first AR supermarket chain in the world.

Each of these supermarkets will have a floor space of around 1.200 m2 and will be located in ‘blank’ public spaces (e.g., train or subway stations, parks, and college campuses). While the naked eye will just see empty floors and walls, people using an AR-capable device will see a complete supermarket, with shelves filled with digital representations of real-world products. To buy products, the user just scans each product with their mobile device, adding it to their online shopping cart. After completing their AR shopping tour, the user receives delivery of the products to their home. This is an enhancement of similar concepts such as the QR-based Tesco supermarkets in South Korea’s subway stations.

 


Figure 25:
Infinite Yihaodian; Source: Augmented Reality Trends

Part 2 :Enhanced Virtual Reality (AR) Technology in Logistics


Nguồn: DHL report

What is the supply chain link that both Walmart & Amazon are lacking?

What is the supply chain link that both Walmart & Amazon are lacking?

“Amazon vs. Walmart” has become shorthand for the competition between online and offiine retailers, yet those two channels are likely to become one as brick-and-mortar retailers go online, and e- tailers seek a closer relationship with consumers. By Bob Sperber

Fortune reported in February that Amazon captured 53 percent of retail e-commerce growth last year to net 43 percent of U.S. online revenue.
That research finding from Slice Intelligence, further revealed a key Amazon advantage: “The average Amazon package was delivered in 3.4 days, compared with 5.6 from everyone else.”

Online competition has led to hundreds of store closures by brick-and-mortar retailers, who continue to retool everything from in-store brand experience to mobile apps and overall e-commerce strate

 

More of their stores offer custom orders, better service to customers who place orders online through their own channels and even orders placed through Amazon.

For its part, Walmart has invested heavily in North American manufacturing, and forged alliances with Facebook, Uber and Lyft to test and expand options for mobile device-based grocery ordering and delivery.

As early as 2014, MIT Center for Transportation and Logistics’ Dr. Yossi Sheffi predicted that Walmart will win the rapid (same-day) delivery game and beat Amazon. One key reason: the brick-and-mortar giant’s thousands of “small warehouses – they’re called stores.”

Amazon is competing online and off. The company Amazon is operating several brick-and-mortar bookstores with more rollouts planned throughout the year.

Next up: grocery stores that eliminate check-out lines by using sensors to automatically charge shoppers as they pick products and walk out the door. Upstream in the supply chain, investments and/or plans range from warehouse robots to cargo planes and truck trailers to a 2016 patent on blimp-like Airborne Fulfillment Centers to launch drones.

“Walmart will win the rapid (same-day) delivery game and beat Amazon ”

Dr. Yossi Sheffi, MIT Center for Transportation & Logistics

The omnichannel line-blurring between traditional and online retailers will bring new technologies to the fore, but none will address problems both share in common: a lack of data visibility to more efficiently and effectively plan, produce, customize and deliver real products to real people. At present, most supply chains are lacking the critical functionality to do so.

What’s Missing? The ‘Perfect Order’

Upstream of the retailer (or e-tailer) are vast collections of brands, manufacturers, contract manufacturers/packagers, logistics firms and many more suppliers.


Both ERP and WMS systems lack the specificity to properly manage major order execution and fulfillment in the age of mass customization, which increasingly entails manufacturing and packaging outsourcing.Leading brands long ago adopted enterprise resource planning (ERP) systems, while their suppliers have largely failed to gain usable data in the warehouse management systems (WMS) they use to produce (and customize) products, resorting to oft-cobbled spreadsheet solutions.

Leading ERP systems are based on the time-honored SCOR Model for characterizing production operations. (APICS pros know the hierarchy well: Plan, Source Make, Deliver, Return; and make-to-stock, make-to-order, etc.) One sub- discipline called “perfect order” has been on the backburner, with proponents waiting for it to become a more practical reality.

One proponent, Dave Blanchard (of IndustryWeek and Material Handling & Logistics), stressed the importance of the perfect order concept as a metric to improve order execution and fulfillment in his book on supply chain best practices. In short, the concept is based on the percentage of error-free steps throughout the life of a purchase order – right product, right condition, delivered on time and damage free, and so on.

The problem is that ERP and WMS systems lack the functional specificity to track operational data at contract packaging and related outsourced locations. Without the right technology to fill the functional gaps, it’s impossible to measure and improve in order to become “more perfect.” SCOR to ongoing work at WERC, the Warehouse Education, and Research Council, we have hope for the further advancement of the perfect order discipline.

As the well-known consultant, Forbes contributor and deep thinker Steve Banker wrote in Logistics Viewpoints, the perfect order metric “is one of the most critical metrics in fulfillment,” and bears close attention.

Brands and their manufacturing and packaging outsourcing partners can use the specific features in solutions such as Nulogy’s PackManager to pursue perfect order-type metrics and key performance indicators.

Rather than target too many variables, experts advise users to start with a few key metrics such as on-time- in-full delivery, correct invoice, and damage-free rates to gain a foothold on optimal execution.

In addition to internal improvements, companies can then target greater integration with partners. When deployed across the operations of multiple supply chain partners, Nulogy says the goal becomes that of achieving a Perfect Order Network™.

 

“The perfect order metric is one of the most critical metrics in fulfillment ”

 

Steve Banker, Vice President, SCM at ARC Advisory Group

 

The goal is to achieve a unified, collaborative enterprise for last-mile product customization with reductions in waste on a global scale for service to retailers and end-consumers.

The days are long gone when “brand” is synonymous with “manufacturer,” and the technology has advanced to the point where today, it is possible to create a transparent and extended enterprise capable of unifying brands and their last-mile service providers.

Without such integration, the kind of market velocity desired by the Amazons and Walmarts of the world will be hobbled by inefficiency. But when brands and their suppliers share the same data – a single source of the truth – new opportunities emerge.

Of course, new opportunities entail new challenges, including those of giving retailers – online and off – greater data visibility.

“Some large retailers have a far more demanding view of what a perfect order is,” Banker says, adding that “a manufacturer must now collaborate with its retail partners to ensure strong in-stock performance at the retail shelf, which ultimately leads to increased revenues and profits for both parties.”

About the Author Bob Sperber

Bob joined Nulogy following 30 years of experience as a business, industry and technology media writer and editor. His work with media outlets and marketing clients have imparted in him a solid understanding of the changes, challenges, and opportunities relevant to many sectors of the global economy.