BAG LOG Issue 14 March 2018 RECLAIM ON DEMAND page 2 CRISBAG EXTENDS TO SELF BAG DROP page 3 NEW BAGGAGE HANDLING SYSTEM FOR MOSCOW SHEREMETYEVO page 4 CRISBAG AND COMPLIANCE: TSA AND CATSA Page 5 MACHINE LEARNING FOR IMAGE RECOGNITION page 8 TREND-SETTING ACCURACY AT CHANGI S NEW TERMINAL 4 The start of operations for the new Terminal 4 (T4) at Singapore s Changi Airport has revealed new self-service technologies and a CrisBag tote-based baggage handling system. These are helping Changi to improve passengers travel experience as well as increase operational efficiency and manpower productivity for airlines and airport agencies. Self-service and automation Changi Airport s new T4 features self-service and automated options offered at check-in, bag drop, immigration clearance and departure gate boarding. Highly redundant CrisBelt transport lines, together with a hold baggage screening system, will ensure safe and efficient baggage operation. All subsequent sortation is made using the latest-generation CrisBag system to ensure the highest level of availability, the shortest transport times and 100% track and trace capability. The CrisBag system integrates an early baggage store so that passengers who arrive at the NEW INSTALLATIONS terminal early may immediately check in their bags and move through to the terminal s airside areas. Changi s CrisBag system spans 5,000 linear metres and is designed to provide the highest level of availability combined with the shortest transport times. Its 100% track and trace capability will also help to increase both security and efficiency. Continued on page 2
2 NEW TECHNOLOGY Continued from front page Controlled by BEUMER Group s software suite, the system allows Changi s Terminal 4 to process 5400 bags per hour. This additional capacity means that the airport can handle 16 million passengers per year through Terminal 4, taking its total capacity to 82 million passengers annually. Rigorous testing Of course, the key to ensuring a smooth and fault-free start to the operation of these automated services is testing. Before the opening of the terminal BEUMER Group processed over a million bags in addition to the training for all ground personnel. Changi Airport Group (CAG) also conducted over 100 trials involving 2,500 airport staff and 1,500 volunteers during the Operation Readiness and Airport Transfer (ORAT) period ahead of the opening of the Terminal. Although the collaboration between BEUMER Group and Changi Airport is based on the latest baggage handling technologies and innovations, it is a partnership which has a strong legacy. Klaus Schäfer, Managing Director, BEUMER Group explains; The original 2002 contract for a tilt-tray sorter in Changi s Terminal 2 was followed by the installation of a CrisBag system in Terminal 3. This operates as a dual baggage sorting system and an inter-terminal transfer baggage system between Terminals 1, 2 and 3. The latest CrisBag, system in Terminal 4 continues the partnership and confirms the shared commitment to using state-of-the-art technology to increase efficiency. Reclaim on demand The arrivals hall is perhaps one of the most frustrating parts of the passenger experience. Once they have disembarked, passengers are expected to wait patiently for their bags to appear on the reclaim carousel. It is one of the few areas of the baggage handling process which has not been the focus of intense improvements in efficiency. Until now. In future, passengers will not have to passively wait for their bags to appear, they will actively be able to decide when to collect their bags from Reclaim on Demand kiosks. The passenger will be able to pull their baggage from an automated reclaim system in the same way that the self bag drop automates the check-in process. The advantages of Reclaim on Demand go much further than giving passengers more flexibility: the number of bags reclaimed by the wrong passenger can be reduced to zero. The airport will be able to determine custody of every bag during the last phase of the baggage handling process in compliance with IATA Resolution 753. Extending 100% traceability to the reclaim area will also eliminate the relatively high percentage of missing-bag reports and the cost to recover and reunite passengers with their lost luggage. Automating the handling of arrivals baggage could support faster baggage transfers too. Loading the in-coming bags into a tote would allow the connecting aircraft to receive the bag tag information as soon as it arrives at the airport. A further benefit is that Reclaim on Demand could drive more revenue to the airport by encouraging passengers to spend more time in the arrivals retail area. Airports could use Reclaim on Demand for all passengers or they could choose to reserve it as a special service for premium passengers. Reclaim on Demand and IATA 753 Loading bags into individual totes as they are unloaded from the plane provides the verification required by IATA Resolution 753 for the transfer of custody from the airline to the ground handling company. The automated scanning and documenting of baggage receipts will make a significant contribution to reducing lost baggage. Not only will fewer bags be lost on arrival, but those which are reported as lost will be easier to trace because each bag will be in its own individual tote.
NEW TECHNOLOGY 3 CrisBag extends to self bag drop Extending tote-based baggage handling to the self bag drop process also extends the reach of the 100% traceability within the Baggage Handling System (BHS). Today s advanced baggage handling technology is mature enough to automate the check-in process and check that each bag is placed securely into its CrisBag tote. The automated system also checks that each bag is within the system s weight and size limits. The use of totes makes it easier to ensure that all parts of the bag, including the straps, are transported safely while also minimising the risk of straps becoming tangled and causing jams in the system. Self bag drop also prevents bags which have unreadable tags from entering the BHS. This eliminates the delays caused by diverting unreadable tags to manual encoding stations and means that the system can be designed without top-loaders. Technology located at check-in detects if a bag needs to be repositioned or if a strap has been left outside of the tote. This requires only minimal intervention from floorwalkers at the check-in and avoids the need for operators to watch for mis-loads when the bag is on the BHS. The advantages of extending the automation of the BHS to the check-in process are clear: all bag and tag data will be automatically registered as soon as the bag is placed into the tote. As each bag remains in its own individual tote from check-in to makeup, each bag is fully traceable at any time. Extending automated baggage handling to check-in. What are the benefits? CrisBag s ability to handle standard and oversize totes on the same system eliminates the need to separate conveyable and non-conveyable baggage at check-in. When the system detects that an item is too large for a standard tote, the system simply sends an oversize tote to the bag drop station. For most airports, Out-Of-Gauge (OOG) represents 5% of the total baggage which is processed. Whilst this is a relatively low percentage, OOG bags demand significantly higher resources of manpower and time compared to standard baggage. The integration of OOG bags into the automated check-in process therefore makes a considerable increase in the overall efficiency of the BHS. Replacing manned check-ins with self bag drop provides a faster, smoother and more flexible service for passengers. Rather than one operator manning each manual check-in, a single floor-walker can help passengers at five self bag drop stations. This typically makes the check-in process up to three times faster for passengers compared to queueing at a conventional check-in. Self bag drop is more space-efficient than conventional manned check-in desks. This allows more space to be allocated for other uses such as advertising. It may even be possible for the bag drop to be wrapped in a design which is customised for the terminal. The technology used in the self bag drop is both tamper proof and security approved.
4 NEW INSTALLATIONS Early 2018 FIFA World Cup victory: On-time completion of Sheremetyevo Airport inter-terminal BHS As sports teams, officials and spectators arrive in Russia for the 2018 FIFA World Cup they will discover a new experience beneath Sheremetyevo Airport: two new tunnels which allow passengers, and their baggage, to move easily between the North and South Terminal Complexes. As Russia s largest airport, Sheremetyevo carried over 40 million passengers on international and domestic flights to more than 200 destinations in 2017. The new North Terminal Complex will increase the hub s capacity to 40 million passengers per year, in line with its Master Plan to increase total passenger numbers to 63 million by 2030. Built directly under the airfield, the tunnels are the first to be built under existing runways. One tunnel will carry passengers whilst the other will transport baggage on a BEUMER autover independent baggage carrier system. In February 2018 BEUMER Group successfully passed the acceptance tests for the Sheremetyevo Tunnel Project as scheduled. The autover system covers 4.8km and carries 187 autoca intelligent Destination Coded Vehicles (DCVs) which each transport a single item of baggage. This provides the airport with 100% traceability of each bag as it transports up to 900 bags per hour, in each direction, between terminals. The speed of the autover system will help to ensure short connection times and future-proof the airport by offering the capacity to support its future growth forecasts. The development of the North Terminal Complex at Sheremetyevo Airport represents a large-scale and unique project for Russian civil aviation and the underground terminal-to-terminal tunnels will help to establish DESIGNED TO FIT Sheremetyevo as a major hub and will raise Moscow s profile for transit, tourist and business passen gers. In Sheremetyevo the BEUMER autover will help keep baggage connection time between terminals to a minimum, while allowing each item of baggage to be safely transported through a tunnel connection and allocated to the correct make-up position. The tunnel is 1300 m long and the autover system runs up to 30 km per hour. With no moving parts, the 4.8 km autover rail will virtually eliminate BHS maintenance within the tunnels. The one item of baggage per autoca principle provides 100% tracking and traceability to ensure optimum baggage handling security, efficiency and throughput.
BHS COMPLIANCE 5 CrisBag and compliance: TSA and CATSA The first tote-based baggage handling system to be certified by the U.S. Transportation Security Administration (TSA) will be installed as part of the Terminal 1 Project at San Francisco International Airport (SFO). The CrisBag Individual Carrier System (ICS) is also the first tote-based baggage handling system to go into operation at a U.S. airport. The TSA has stringent design and delivery specifications for airport baggage handling systems and the BEUMER Group team worked diligently to meet these planning guidelines and design standards, explains Klaus Schäfer, Managing Director, BEUMER Group A/S. Throughout the compliance process, the TSA has been extremely supportive of the tote-based Individual Carrier System (ICS) and Radio Frequency Identification (RFID) technologies used in CrisBag. Last year, BEUMER Group became the first vendor to achieve TSA certification for in-tote baggage screening. Crucially, the integration of the Smiths Detection CTX 9800 DSi Explosives Detection System (EDS) enabled CrisBag to meet the detection and false-alarm rates mandated by the TSA specifications. The TSA tested the effectiveness of the CrisBag system across a range of real-life conditions and situations at its full-scale Transportation Systems Integration Facility near Washington, D.C. The elements of the CrisBag system which were assessed by the TSA included tracking, throughput, imaging, sorting and the detection of missing, unknown and oversized bags. This in-depth evaluation also recorded how rapidly the CrisBag system associated specific bags with an individual tote. TSA certification will strengthen the business case for airports to adopt the common-use model of an RFID-enabled, tote-based baggage handling system. The integration of an RFID tag into each CrisBag tote enables virtually 100% traceability and allows the airport to track the location of every bag within the system at any time. CATSA pre-border compliance CrisBag is also helping airports to comply with the Canadian Air Transport Security Authority (CATSA) requirements for US transborder clearance. For Calgary International Airport and Ottawa International Airport, this is a key advantage of using the CrisBag system. Bags are placed into individual CrisBag totes at check-in and remain in the same tote to ensure the 100% traceability which is required to comply with the CATSA pre-border clearance process. The tote system also allows efficient manual inspection when required which adds to the fail-safe capabilities of the baggage handling process. Of course, compliance to the CATSA pre-border specification was not the only factor in the decision to install a CrisBag system. Both Calgary and Ottowa had other strategic objectives which needed to be considered before they upgraded their baggage handling systems to CrisBag. Both airports were also looking for a reduction in energy consumption and an increase in the efficiency of the overall baggage handling process. These objectives led them away from using a conventional conveyor-based baggage handling process in favour of the CrisBag system which they have installed in their Domestic, International and US transborder terminals. Airports of all sizes are embracing new baggage handling technologies to deliver facilities for US border preclearance, comments Moritz Bender, Director Sales & Business Development, Airport, BEUMER Group. The modularity of the CrisBag system allows the baggage handling system to be designed to fit each airport s individual objectives and specifications. BEUMER s track record of long-term partnership with airports means that we will also ensure that the baggage handling system has the flexibility to respond to future changes in passenger numbers as well as new regulations.
6 BIG DATA Turning Big Data into a digital service will be useful and valuable to us. It is like a museum curating an exhibition of art: not every piece will be used because the museum will select only the pieces which fit the context in which they will be displayed. harnessing the data to help all airports with CrisBag systems become smarter. Also, the data is local and does not have an identity in the cloud. This eliminates the risk of any data being compromised. An interview with BEUMER Group Business Development Director, Per Engelbrechtsen. What is Big Data? Throughout the airport baggage handling industry, 80-90% of all log files can be considered dark data. Vast amounts of data are collected, but only a small percentage is used to contribute to the decision-making process. Imagine what we could achieve if we could find a way to harness the power of the sea of data that flows into each business. Today, Big Data is a buzz word, but BEUMER Group is investing the resources to transform that buzz word into value creation. The questions that we must ask are: Why do we want data and where is the value in data? Data is only valuable when we understand the context in which it has been collected. BEUMER Group has explored these endless pools of data to find the trends that can help us understand the information that provides value to a business. What happens to the data? As the data is collected, it is curated so that it can be brought back in the correct context to reveal a trend. We need to establish why the data is being collected; what it is being used for, and the context in which it will be used. Once we establish this, we can curate the data that we believe How can we use Big Data? Computing power is the other aspect of the Big Data revolution. Computing power is all around us: it is in our phones, tablets and many of the systems that we use every day. Can we use the processing power that we find all around us to turn data into value? Take the CrisBag system for example. Sensors are used to collect data from every single section of the baggage handling system. Now, consider the fact that a tote travels through the CrisBag system much faster than a person could visit every part of it. So, can we make these sensors work harder for us? Could the tote become a Mars traveller that collects data, so we don t have to go and check the system for ourselves? Can we make use of the tote s computing power to do analytics as it travels around the CrisBag system? If we do this in several locations, we are So how does Big Data support Machine Learning? Big Data is also the basis for machine learning. The machine must be taught how to extract information, and how to recognise and understand what the language means. The machine must absorb the information and apply it in context. It must also be able to build a hypothesis for humans to decide what the data tells us. This is when machines and humans work together. Humans have greater cognitive capabilities than machines but, unlike machines, we cannot go through millions of lines of data, such as pictures or text, without getting tired and making errors. A system can do that if it is taught what it needs to look for. As human experts, we can delegate some of these tasks to Artificial Intelligence, sharing the work-load so that we can broaden our own capabilities and those of our businesses.
REMOTE ENCODING 7 The transfer bag challenge Timing is crucial for many airport operations, but it is particularly important for the efficient handling of transfer baggage. An analysis carried out by BEUMER Group revealed that, within an overall connection time of 45 minutes, the BHS has just seven minutes to process each item of baggage. The typical times taken for each element of the baggage handling process are: 18 minutes: Aircraft to transfer belt Seven minutes: BHS processing 20 minutes: Chute to aircraft With such tight timeframes, it is easy to see how delaying a bag by just a few minutes could mean that the bag misses the connecting flight. A fully automated process can usually handle most baggage within the typical timeframes, however, any bag with an unreadable Bag Source Message (BSM) tag will need additional processing time. Typically, this is around 2-3% of all baggage. The delay in processing a bag with an unreadable tag is caused by the need to re-direct the bag to a Manual Encoding Station (MES). Here, an operator would manually enter the BSM information before re-inducting the bag back onto the BHS. The manual encoding process can take several minutes which can mean that the bag misses its connecting flight. It also means that each manuallyencoded bag will touch the sorter twice, impacting on the overall capacity of the sorter. The video coding system solution The introduction of a Video Coding Systems (VCS) can eliminate the need to divert unreadable bag tags for manual encoding. Located at strategic locations on the sorter, VCS cameras capture an image of each bag tag. The image is then sent, from the web-based application, directly to the MES operator who uses a touch-screen to zoom-in to view and manually encode the tag. Typically, the installation of VCS reduces the workload of the MES by up to 50% and eliminates the need for operators to physically turn or lift the bag. The reduction in MES redirects can be highly effective in reducing the number of short-shipped bags in addition to cutting costs for the airport and airlines. It will also help to shave peaks and minimise bottlenecks by enabling bag-source information to be encoded while the bag is transported inside the BHS. In addition, the VCS adds to manpower flexibility by allowing control-room staff to supplement or relieve MES staff during coding peaks. One airport which already uses the BEUMER Group VCS, has reported that it prevents up to 10 shortshipped bags a day. This saves an estimated 100 per bag on the cost of reuniting each bag with its owner in addition to making a significant contribution to passenger satisfaction.
8 MACHINE LEARNING The Machine Learning solution Another approach to solving the issue of unreadable bag tags is to use Machine Learning. BEUMER Group is implementing Machine Learning for image recognition by adopting a cloudbased technology, based on a pre-trained Neural Network. BEUMER Group s Neural Network has been pre-loaded with many thousands of images of read and no-read bag tags. These images allow the system to learn to identify correct and incorrect tags. In a live system the network receives 120 images of each bag captured by six cameras located on the baggage handling system. The Machine Learning software instantly recognises whether the image is of a tag, sticker or belt before selecting and sending the best images to the operator. The operators receive these images on tablets. This means that the software not only eliminates the need to divert bags to the MES, but also frees operators from having to stand at a fixed manual encoding workstation. Consultation with a group of international airports, BEUMER Group has revealed that the decisions made by neural networks are more accurate than those made by human operators. Man-made decisions are not always precise because humans cannot match Machine Learning s ability to process every factor in a single moment. Initially, Machine Learning will be used to support operators in making faster decisions. Eventually, however, this technology could have the capability to make decisions. The probability is that those decisions will be more consistent and accurate than those made by their human counterparts. Machine learning Machine learning is a particular approach to artificial intelligence (AI) which is the branch of computer science concerned with making computers behave like humans. Some of the hottest AI technologies right now include natural language generation, speech recognition, virtual agents and robotic process automation. In its Video Coding System, BEUMER Group processes images through machine-learning software which instantly recognises whether the bag has a tag, sticker or belt. The software then selects and sends the best images to the operator for remote, manual encoding. The reason for using Machine Learning is that, whilst humans like to feel that we are infallible, it is the machines which are capable of repeatedly making the most consistent and accurate decisions. BEUMER Group A/S P.O. Pedersens Vej 10 DK-8200 Aarhus N Denmark Phone +45 87 41 41 41 www.beumergroup.com/airports