Crane Productivity Improvement

Alex Goussiatiner P.Eng., M. Sc. Senior Container Terminal and Transportation Specialist Sandwell Engineering Inc. e-mail: [email protected]

  Systematic Approach to Quayside Container Crane Productivity  Improvement  Introduction  Besides the terminal reliability factor (the fact that the vessels always sail on time) and terminal tariff rates, gross rate of the quayside crane is the most important factor taken into account by shipping lines when evaluating terminal services. The gross crane rate is a comprehensive indicator of vessel operation performance as delays under the control of the terminal operator and shipping lines are not subtracted. Most of the terminals and shipping lines use the following definition of gross crane rate: “the total containers handled divided by the allocated crane time.” Allocated crane time is the total number of working crane hours. In contrast, the net crane rate is the total containers handled divided by the elapsed crane time, where elapsed crane time is the allocated crane time less operational delays. The gross crane rate directly affects time at berth. For instance, if a vessel is working with three cranes and exchange volume is 1200 cont, increasing gross rate from 27 cont/h to 30 cont/h reduces time at berth by 1.5 hours. Reducing time at berth gives a competitive advantage to the terminal and in most cases leads to increase in volume and revenue. At the same time, increase in the gross rate improves overall efficiency of the operation as it allows the terminal to reduce allocated crane hours and save on the costs associated with manpower, maintenance, etc. Using the terminology of the economical value analysis, gross crane productivity is a “basic function" for the terminal services. In other words, the gross crane rate is anything that makes the service sell and the terminal survive. The current volume downturn highlights the situation. An increase in crane productivity should help terminal operators retain and increase volume. Performance benchmarking of various modern container terminals suggests that significant productivity reserves exist in most of the container terminals worldwide as container terminals with similar quayside cranes and types of operation produce very different results.

This paper discusses what technical and operational options are available to terminal operators for increasing productivity of existing cranes. It also describes how statistical and simulation modeling can be used for prediction of gross crane rates after implementing certain solutions.

This paper does not promote any particular technology or solution, rather our focus is to develop a systematic and comprehensive engineering approach for crane productivity and improvement projects.

Factors Affecting Productivity  There are three key characteristics of the crane performance: cycle elapsed time, lifting spectra and operational delays which determine the gross crane rate. The following analysis presents the underlining factors affecting each of those characteristics:

Cycle Elapsed Time   The Cycle Elapsed Time (or net cycle elapsed time) represents elapsed time required to perform one crane cycle in normal working condition when there are no operational delays. The following factors affect the crane cycle elapsed time: Type of Operation. Type of operation includes a combination of the direction (loading or unloading) and location (deck or hold) of the operation. The type of operation defines what activities are performed during the cycles. For instance, as figure 1 shows, unloading from a deck requires twist locks removal which is not required for the loading to hold (see fig.2).

Figure 1 – UD (Unloading from deck) activities

Figure 2 - LH (loading to hold) activities

Typically, if operation is performed on deck, it requires shorter Crane Cycle Elapsed Time. Also loading operations usually require more time than unloading. Cycle Type. By the cycle type we understand combination of the lift mode and container size: 20-sigle 20’ lift, 40-single 40’ lift, 20-20- twin 20’ lift, 40-40 – tandem lift (two 40’ alongside), 40|40 vertical tandem lift (two 40’), etc. Cycles with a larger number of containers always require longer crane cycle elapsed time. Crane Speed and Lifting Capacity. Hoisting, lowering, lifting capacity and traveling speed of cranes affect cycle elapsed time. For example, newer super Post Panamax cranes almost always outperform the old Panamax crane in the cycle elapsed time. Yard Interface Efficiency. Even during normal operation, cranes can have short ‘waiting’ periods resulting from tractor-trailer positioning, absence of the ground slots for containers (ground operation), delays in the tractor-trailer arrivals, etc. Usually, those ‘waiting’ periods are not registered as operation delays and ‘blended’ inside the cycle elapsed time. The main reason for waiting quayside cranes is either low number of the prime movers or large cycle time for the prime movers. Large cycle time depends on traveling distances, number of yard cranes used for the operation, yard cranes workload, traffic congestion, etc. In the tractor-trailer mode, the tractor-trailer systems are used to move containers to and from the yard.

In the ground mode, gantry cranes ground containers during discharge and pick directly from the ground during loading. The straddle carriers or shuttle carriers are used in the ground mode to move containers to and from the yard. In most cases, the ground mode provides better performance than the tractor-trailer mode. The operation in this case is decoupled in time and the gantry is less likely to have to wait for the tractor-trailer and vice versa. Vessel Presentation. In most cases, each vessel service has its own vessel presentation or pattern, which remains persistent for all vessels calling on the terminal for the service. The presentation prescribes the main dimensions of the vessel operation: distribution of the containers between bays, stowage patterns inside the bays, degree on which containers are ‘scattered’ over the vessel, ratio between 20’ and 40’ bays, etc. Quayside cranes always do better on a well presented ship than on a poorly presented ship as they are required to visit every deck for a handful of containers. The vessel presentation affects Cycle Elapsed Time. For instance, the Cycle Elapsed Time is shorter when the whole bay is discharged at once (well presented vessel) instead of working ‘blind’ in center of the bay (poorly represented vessel).

The cycle elapsed time is also affected by temporal factors related to the wind and wave conditions. Wind Conditions. A high wind load increases spreader swaying of the container. As a result, more time is required for spotting of the container. Wave Conditions High waves, especially long period waves generated by ocean swells, cause increase in vessel motion for the vessel moored at berth. A combination of different vessel motions caused by waves increases Cycle Elapsed Time and decreases the crane rate. For instance, the sway and surge motion combinations caused by long period waves makes it harder for crane operators to spot a container inside the ship and enter the cell guides.

Labor Motivation and Competence Most terminal operators cite lack of motivation and lack of proper training as factors in lessened productivity. Low motivation of terminal operators can often be attributed to a lack of interest in quality or final results. Inadequate proficiency of the crane operator and other workers can be caused by the dexterity factor and lack of specialized training. Specialized training using mechanical and computer simulation brings improvement in proficiency and also allows objective selection By itself, the proficiency factor cannot be over estimated. Data shows that an excellent operator has an advantage of approximately 10%-15% in the Cycle Elapsed Time over an average operator.

Lifting Spectra  Suppose that we have portioned all crane cycles into groups according to the cycle type. The array of cycle counts will represent the lifting spectra:

Cycle Type

20

40

20-20

40-40

Number of Cycles

S1

S2

S3

S4

The ‘strong’ lifting spectra is defined as the spectra with a greater number of multi-container cycles and a greater average container count per cycle. The ‘weaker’ lifting spectra produces a lower average container count per cycle. Container Spreader Type affects the lifting spectra. Most of the cranes in production are equipped with telescopic spreaders capable of handling single 20, 40, 48 foot containers. More and more cranes are equipped with the twin lift spreaders capable of handling two twenty foot containers at once. Smaller number of cranes are equipped with the tandem lift spreader, which is capable of handling two 40 foot or four 20 foot or their combinations. Crane Lifting Capacity also affects the Lifting Spectra. For instance, most of the tandem-lift cranes in production have lifting capacity under spreader lower than 120 t. As a result, the cranes often cannot handle four full containers at once. An increase in the lifting capacity improves usability of the crane in the tandem lift mode and makes the lifting spectra stronger. Vessel Presentation, ratio between 20 and 40 foot container, ratio between full and empty containers on board affect the Lifting Spectra as those characteristics control feasibility of the multi container cycles.

Operating Delays  Operating delays represent situations when vessel operations cannot continue based on reasons under the control of the terminal operators or otherwise. Operating delays are defined according to the classification adopted in individual terminals: •

Unlashing delays involve work done by the lashing crews who remove lashing bars and unlock twist locks. Unloading from bays on deck cannot start until the lashing crews are finished.

•

Hatch cover handling delays relate to removing and setting back the hatch covers.

•

Twist locks delays are caused by failure to unlock twistlocks, false unlocking, damaged twist locks, etc.

•

Wrong/bad container delays are caused by errors in container location, number, etc.

•

Traveling between bays delays

•

Crane downtime delays are caused by crane and spreader breakdowns.

•

Adverse weather, wind or waves conditions are delays caused by bad weather, wind and wave conditions.

•

Others: all other delays including spreader changes for over dimensional containers, problems with the terminal operating system, adverse weather delays, etc

There are a number of factors affecting operating delays such as: Labor Motivation and Competence: motivated and competent workers resolve problems quickly and make fewer mistakes in container checking, etc. Vessel Presentation also affects operation since poorly presented vessels require more traveling between bays and more hatch cover handling.

Productivity Improvement Projects: Step‐by‐Step  We suggest that improvement processes should follow these stages:

Recognize Problems and Goals  From our experience, most terminal operators can identify the weakest elements in vessel operations. But to identify desirable levels of crane productivity that make the terminal competitive compared to others in the region, we recommend market analysis. This analysis should determine the minimum and desirable performance for servicing certain ‘targeted’ vessel services.

Propose Solutions or Methods to Achieve Goals (Synthesis)  Various solutions are known to improve crane performance. Cost is associated with each type of solution. Some solutions require capital investments such as upgrading crane infrastructure, implementing automated technologies, etc. Others do not require capital investment, but still require financing. These include personnel training, manpower motivation, reengineering of the technological process, etc. At this stage, a solution or combination of solutions can be implemented.

Validate  In addition, any adapted solution should comply with all of the following constraints: Feasibility: Not all of the well known solutions are feasible in particular terminal settings. For instance, installation of the tandem-lift spreader increases the wheel load. Thus structural engineering analysis of the rail and wharf structures is required before the solution can be adopted. The feasibility studies narrow the decision down to a few viable options. Safety: The solution should comply with all the safety regulations at the terminal and improve the safety for the workers.

Environmental impact: The solution should be in compliance with all environmental regulations. Flexibility: Unexpected changes in working schedules, handling cargo and volumes should be accommodated in the decision. 

Evaluate Productivity Gain and Assess Risks of the Solution (Analysis)  In this step, we estimate an increase in productivity as a result of implementing the solution. There is also the risk that the solution might not bring what is expected. The goal is to identify a set of options, with each option providing the targeted productivity gain.

Select the Solution  In the end, the most cost effective option should be accepted. To determine which option is the most cost effective, we recommend creating a financial model based on the net present value method. The model will use the startup and operational cost as the inputs.

Implement the Best Solution  Proper project management is required at the planning and execution phases of the productivity improvement projects. This effort requires flexibility and willingness to optimize the chosen solution. Special attention is required for maintaining productivity gains. A well organized improvement program with wide participation from all workers is instrumental to achieving goals.

Possible Solutions  The ‘causal diagram’ presented in Figure 3 depicts the solutions and productivity factors (variables), discussed in this paper. The diagram consists of arrows connecting variables, things that change over time, to show how one variable affects another. The black arrow means that the first variable causes a change in the opposite direction in the second variable and the red arrow means that the first variable causes a change in the same direction. For instance, the ‘vessel motion’ variable is connected to ‘cycle elapsed time’ via a red arrow, as increase in vessel motion always leads to an increase in crane cycle time. In the diagram, the green variable depicts ‘operational’ solutions, which are based on the optimization of the operational process. The yellow variables depict technological solutions. The causal diagram is a kind of systems thinking tool, which helps to identify the ‘causal’ relationships between factors.

Operational Solutions – ‘Lean’ Thinking  Lean thinking is a systematic approach to business processes with the aim of doing more with less while still providing customers with exactly what they want. While lean thinking is not based on a certain technical solution, costs related to planning and executing the changes and costs for modifying the terminal operating systems should be considered.

Eliminating Redundancies Activities in the working cycles depend on rules that were adopted in each terminal a long time ago. These rules are not performed in all the terminals all the time; for instance, recording of the ‘seal’ number is not mandatory in all the container terminals. These redundant activities can be eliminated after achieving an agreement with the shipping lines. Double Cycling During the operation in hold, container cranes can unload an import container and return to the hold with an export container. This reduces empty crane travel, but increases cycle elapsed time. In addition, it requires the yard cranes in both export and import stacks to work simultaneously. This reduces the yard interface efficiency and increases the ‘waiting’ time. Overall, according to published test results, the double cycling increases cycle elapsed time by approximately 60% in comparison with the single lift. Common Pool Over the years it has been proven that combining prime movers allocated to a number of the vessel cranes into one group increases utilization of the fleet and improves the efficiency of the yard interface. This reduces the ‘waiting’ time for the crane.

Robotized Technologies  Container Number OCR Digital image recording and optical character recognition software allow identification and recording of the containers handled by the crane. This eliminates the need for the manual recording during the crane cycle and reduces delays related to the bad/wrong containers. Automatic Ship Mooring Systems. A number of equipment manufacturers have patented various mechanical devices which allow securing of large container vessels alongside the berth. A prime target for the systems is to reduce time for the vessel mooring, but at the same time the mooring systems reduce various vessel motions by absorbing the energy. Such reduction of the vessel movement prevents operations delays and reduces crane elapsed cycle time. Robotised Twistlock Unlocking. Loxystem (Sweden) introduced Remote Controlled Automatic twist lock system (RATs). This system allows unlocking twist locks for the container on deck upon receiving command from the crane operator. The operator should become aware if there is a failure to unlock immediately. Adopting the system by vessel operators can virtually eliminate the unlashing delays and reduce “twist lock unlocking failure” delays. Also, the ability to disconnect twist locks remotely can lead to wider adoption of the vertical tandem lift operations. Robotised Twistlock Handling. This system allows automatic or semi-automatic removal and insertion of the twistlocks for the containers on deck. For instance, Ram Spreaders (Singapore) produced the PINSMART system where the RAM PinSmart machine comprises a steel platform with the robotic corner handling modules at each corner. When unloading, the machine senses that the container is nearby and the robotic corner handling modules automatically remove the twistlocks. During loading, the worker places a twist lock at each corner handling module. A container or containers can then be lowered onto the machine and the twist locks will

be automatically inserted. The system drastically reduces time for handling twist locks, which is particularly important for the terminals that operate in ground mode, where removing twist locks is an integral part of the crane cycle time.

wind anti-swaying system dual spreader lifting capacity increase dual spreader

vessel motions spreader swaying OCR automatic TL insertion trailer positioning syst. lifting spectra double cycling lifting spectra

lifting capacity increase

automatic twist locks

OCR

unlashing time hatch cover handling TL unlocking failure wrong/bad container crane downtime other delays

# cont per cycle

waves automated mooring system

operating delays

common pool eliminating redundancies double cycling training

cycle elapsed time

Figure 3 - Effect of the improvement solutions

Crane Control Systems. These systems (which are part of the TOS) allow automatic exchange of the data between parties involved in the operation (crane operators, tractor operators, checkers, etc) in real time. As result, the crane operators and tractor drivers are receiving instructions on graphical monitors and less radio contact required. Implementation of such systems in a number of terminals proved that the systems can reduce the cycle time and drastically reduce wrong/bad container delays.

Tandem Lift Spreaders  Originally the tandem lift spreaders were only available as part of the new cranes. However, the situation has changed. At the moment, a number of manufacturers can produce tandem lift spreaders which can be installed in existing cranes if they match the minimum requirements defined by spreader manufacturers. For instance, Stinis (Netherlands) invented and produced split-headblock spreaders that can be installed in cranes with lifting capacity greater than 95 t under the wires (the spreader weight is 12.2 t ). The spreader can be quickly transformed from tandem lift (two 40’ or four 20’) mode to single mode and vice versa.

Evaluating Productivity Gains    Statistical Models and Factorial Analysis  In practice, “net” and “gross” crane rate is calculated from computerized or manual time recording of the crane operation and operating delays. The recording is usually done for the particular operation at the bay. For instance, figure 4 represents recording of the unloading operation from bay 12D. The first column is used for recording Elapsed Crane Times and Delay Times in minutes. Elapsed Crane Time includes elapsed time when the crane was performing actual movements and ‘waiting’ time which represents waiting for the prime movers or containers to be handled. Delay Time represents elapsed time when the crane was not performing cycles and was waiting until the problem was resolved. The number of containers handled is recorded separately for each type of lift: single 20’, single 40’, twin 20’, etc Figure 4 - Time Recording Sample Voyage: Crane: Shift: Bay: Operation Type: Operator:

King of the Sea V 005/006 G-08 01:00 12 D UD (unloading from deck) John Smith

The operational recording data can be aggregated using classifiers for the vessel service, crane type (see Fig. 5) . In the aggregated table, the following variables—vessel presentation pattern, crane type, type of operation—are identified as the influencing factors, while the lifting spectra, cycle elapsed time and delays are the response variables. Figure 5. Aggregated Statistical Data

The statistical observations with the same influencing factors are relatively homogeneous and can be used for various comparative studies. For instance, to determine increase in the crane rate after implementing an automatic twist lock system for a certain vessel presentation pattern and crane type, we can first establish a baseline by calculating average gross crane rate, based on all the observations for the Vessel Presentation and Crane Type combination. Then, we can modify the statistical model, and ignore the delays related to the twist locks unlocking (“A” and “C”), and recalculate the gross crane rate that would have resulted with an automatic twist lock system.

Discrete Event Simulation  Statistical modeling is an excellent tool for the estimation of the productivity gains, but it has its limitations. For instance, introducing new equipment types and/or modifying of the operational strategies quite often changes how the parties of the process (crane operator, tractor-trailer operators, checkers, etc) interact with each other. In this situation, there is no relevant statistical data available for the analysis and discrete event simulation is applicable. In discrete-event simulation, the operation of a system is represented as a chronological sequence of events: engaging of the container, lifting, lowering, checking, tractor trailer arrival, tractor-trailer positioning, etc. Each event occurs at an instant in time and marks a change of state in the system. Each activity has its own random variables with predefined statistical distributions for the elapsed time. Statistical modeling is only used to calibrate the simulation model.

Final Remarks    The volume downturn and fierce competition will force many terminal operators to take a hard look at their existing operation and equipment: is it really efficient? What can be done to improve the crane productivity? This will have lasting impact on the container handling technology. Hopefully, this paper can help on the road to efficiency.