Small Ro/Pax Vessel Stability Study

Small Ro/Pax Vessel Stability Study Primary author: Henrik Erichsen, M.Sc., Lead Technical Specialist in Charge, Lloyd s Register EMEA Co-authors: Hans Otto Kristensen, Director, M.Sc. FRINA and MSNAME, HOK Marineconsult ApS, Jørgen Juncher Jensen, Professor, Department of Mechanical Engineering, Technical University of Denmark, AMOS Centre, Norwegian University of Science and Technology, Erik I. Tvedt, M.Sc., FRINA, Special Adviser, Technical Regulation, Danish Maritime Authority. Abstract In 2009 new damage stability requirements for passenger ships based on a probabilistic method were adopted by IMO and are now part of the current SOLAS Chapter II-1 regulations (SOLAS 2009). The mandate from IMO was to keep the same safety level as inherent in the old deterministic damage stability regulations in SOLAS (SOLAS 90). During the rule development prior to the adoption, it was argued that the safety level for large passenger ships should be increased, but small ro/pax vessels were only rudimentarily looked at and small vessels with very high attained index were seen as non-representative. Currently there is a renewed debate in IMO regarding the required damage stability safety level for passenger ships. The damage stability safety level for small ro/pax vessels has also been discussed outside of the IMO assuming that the damage stability safety level for small ro/pax designs is perhaps not sufficient, i.e. that the current safety level according to SOLAS 2009 is less than the old safety level according to SOLAS 90. In order to establish a solid foundation for the discussion, this study was made possible by a grant from The Danish Maritime Fund. The study focus on small ro/pax vessels in a range from 32 m to 100 m in length and 100 to 600 passengers/persons, and the outcome of this study is described in details in this document. INTRODUCTION Five existing vessels currently operating in Denmark have been chosen for the study. The majority of these vessels have been built within the last 17 years and represent a wide range of the different ferry designs operating in Danish waters. Based on the as built safety level (pre-solas 2009 limiting KG/GM curves) the probabilistic damage stability of these vessels, including the application of the new survivability factor s covering damage to the ro/ro decks as agreed at the 55 th Session of the IMO SLF Sub-Committee, but not yet adopted into SOLAS, has been calculated according to the SOLAS 2009 requirements and the attained index is compared to the required index R. The study also investigates the other damage stability requirements of SOLAS 2009 such as the deterministic damages according to Regulation 8 ( minor damage concept) and, to a certain extent, also the bottom damages according to Regulation 9.8. For each of the chosen existing vessels, the paper indicates, on this basis, whether the damage stability safety level is maintained by SOLAS 2009 regulations. The paper also describes which factors are the actual dimensioning criteria for damage stability and how designs are influenced by seemingly unrelated factors like fire insulation etc. SOLAS DAMAGE STABILITY REQUIREMENTS When SOLAS 2009 was introduced 1 st January 2009 the damage stability requirements for passenger ships went from a deterministic to a probabilistic approach. Prior to 2009 the damage stability requirements was either a 1 or 2 compartment deterministic damage stability standard based on the number of persons onboard the ship as well as the criterion number. In SOLAS 74 the damage stability criteria were very limited. This was strengthened in the 1983 amendments and then again with the SOLAS 90 requirements. In addition to these damage stability requirements, the floodable length requirements included in SOLAS, Part B, regulation 4, demanded a high degree of subdivision for those ships. After the Estonia disaster in 1994 some countries implemented water on deck damage stability requirements based on the SOLAS 90 concept. This further strengthened the damage stability requirements. In parallel to the deterministic SOLAS requirements, a probabilistic approach for passenger ships had existed since the IMO Resolution A.265(VIII) was adopted in 1973. Even though this resolution was seen as equivalent to the SOLAS deterministic approach, it was seldom in use. In 1992 a probabilistic damage stability standard for dry cargo vessels was adopted into SOLAS and this initiated the work to harmonize the different damage stability standards used for dry cargo ships and passenger ships. THE PROBABILISTIC CONCEPT The probabilistic damage stability concept is, in theory, very simple. The concept uses the probability of survival after collision as a measure of the ships safety in a damaged condition. This probability is referred to as the attained subdivision index A in the regulations. It can be considered an objective measure of ships safety and, ideally, there would be no need to supplement this index by any deterministic requirements. The philosophy behind the probabilistic concept is that two different ships with the same attained index are of equal safety Erichsen Small Ro/Pax Vessel Stability Study 1

and, therefore, there is no need for special treatment of specific parts of the ships, even if they are able to survive different damages. The only areas which are given special attention in the regulations are the forward and bottom regions, which are dealt with by special subdivision rules provided for cases of ramming and grounding. The attained subdivision index A must be equal to, or greater than, a required subdivision index R which has been politically decided. A R The Required Subdivision Index R The required subdivision index R is stated in SOLAS 2009, Chapter II-1, Part B-1, Regulation 6.2.3: valid for a trim range of +/-0.5 m. If the operational trim of a specific ship exceeds this limit, additional calculations with trim must be carried out. For all the ships calculated in this study only level keel has been used as none of the ships have operation trims exceeding +/-0.5% of the subdivision length L s. The Survivability Factor s The survivability factor s i is based on requirements to heel, GZ and stability range. The factor s i, for any damage case at any initial loading condition, d i, shall be obtained from the formula: s i = minimum {s intermediate,i or s final,i s mom,i } s intermediate,i is the probability to survive all intermediate flooding stages until the final equilibrium stage. 5,000 R = 1 - L s + 2.5N + 15,225 s intermediate,i = GZ!"# 0.05 Range!/! 7 Where L s is the subdivision length and N depends on the number of persons and the provided lifeboats onboard. N = N 1 + 2N 2 N 1 = number of persons for whom lifeboats are provided. N 2 = number of persons (including all crew members) the ship is permitted to carry in excess of N 1. While the subdivision length L s only has a minor influence on the required subdivision index, the index is more influenced by the number of persons onboard and the provided lifeboats. The majority of the selected ships for this study are operating in sheltered domestic waters and are thereby not carrying lifeboats hence N 1 is 0. The Attained Subdivision Index A The attained subdivision index A is based on the probability that only the compartment or group of compartments under consideration may be flooded and the survivability factor s which accounts for the probability of survival after flooding the compartment or group of compartments under consideration. Where GZ max is not to be taken as more than 0.05 m and Range as not more than 7. s intermediate = 0, if the intermediate heel angle exceeds 15º. Where cross-flooding fittings are required, the time for equalization shall not exceed 10 min. s mom,i is the probability to survive heeling moments such as crowding of passengers to one side, a wind force acting in a damage situation and a heeling moment due to the launching of all fully loaded davit-launched survival crafts on one side of the ship. s final,i is the probability to survive in the final equilibrium stage of flooding. It is calculated in accordance with the formula in SOLAS 2009, Chapter II-1, Part B-1, Regulation 7-2.3: s final,i = K GZ!"# 0.12 Range!/! 16 Where GZ max is not to be taken as more than 0.12 m and Range as not more than 16. The factor K is calculated by the formula: A = p! s! K = θ max θ e θ max θ min The attained index A is calculated on three different draughts in order to cover the operational draught range, i.e. the deepest subdivision draught d s (index A s ), the light service draught d l (index A l ) and the partial subdivision draught d p (index A p ) which is the light service draught plus 60% of the difference between the light service draught and the deepest subdivision draught. The total attained subdivision index is summed up as follows: A = 0.4A s + 0.4A p + 0.2A l Where θ min is 7, θ max is 15 for passenger ships and θ e is the equilibrium heel angle experienced in the damage. K is equal to 1 if θ e θ min and is 0 if θ e θ max In order to cover water on deck effects on Ropax ships a new s factor was agreed at the 55 th Session of the IMO SLF Sub- Committee. This new is applied in the final equilibrium for damages involving the Ro/Ro deck. The attained subdivision index calculation is carried out on level trim and that calculation is valid for a trim range of +/-0.5% of the subdivision length L s, e.g. for a ship with a subdivision length of 100 m the level trim attained index calculation will be s final,i = K GZ!"# 0.20 Range!/! 20 Erichsen Small Ro/Pax Vessel Stability Study 2

Even though the new is not in force yet, the consequences for introducing it to small Ropax ships has been analysed in this study. OTHER SOLAS DAMAGE STABILITY REQUIREMENTS To avoid passenger ships being designed with, what might be perceived as unacceptably vulnerable spots in some part of their length, SOLAS 2009, Chapter II-1, Part B-1, Regulation 8 include a deterministic minor damage standard. This standard is based on the subdivision length L s and the number of persons onboard the ship and is, in many cases, the limiting requirement for smaller passenger ships. For passenger ships carrying 400 persons or more, this is in fact indirectly a 2 compartment damage stability standard. Another damage stability standard which was implemented with the SOLAS 2009 regulations was the bottom damages that have to be complied with if the ship is not fulfilling the double bottom requirements in SOLAS 2009, Chapter II-1, Part B-1, Regulation 9. These bottom damage requirements are presented in paragraph 9.8 of the regulation. THE SELECTED SHIPS Five existing vessels currently operating in Denmark have been chosen for the study. The majority of these vessels has been built within the last 17 years and represents a wide range of the different ferry designs operating in Danish waters. The ships are ranging from an overall length of 31.90 m to 99.90 m and from 100 to 600 persons onboard. The majority of the ships were built according to the SOLAS 90 1-compartment or 2-compartment standard. The newest ship from 2012 was built in accordance with the SOLAS 2009 probabilistic standard as well as the deterministic standard of the EC Directive 2003/25 covering water on deck. The oldest ship SRPX06 was built in 1989, i.e. before SOLAS 90 was introduced. This ship has later been upgraded to the SOLAS 90 1-compartment standard and was included in the study to see how the results would turn out for a pre SOLAS 90 ship. This ship has a pronounced B/5 subdivision nearly over the entire length. The general particulars of the selected ships can be seen in table 1 below. Table 1 Main Particulars of the selected ships Ship Built Type LOA Bmld D T Freeboard Pax Damage Stability Standard SRPX01 1999 D-E* 31.90 9.40 3.30 2.20 1.10 100 SOLAS 90, 1-compartment standard SRPX05 1999 S-E* 49.90 13.10 3.70 2.64/2. 70 1.06/1.00 395/250 SOLAS 90, 1-compartment standard SRPX04 1997 S-E* 68.50 15.80 4.55 2.90 1.65 580 SOLAS 90, 2-compartment standard SRPX06 1989 D-E* 95.80 15.00 4.90 3.60 1.30 399 As built: SOLAS 74/83, 2-compartment standard Now SOLAS 90, 1- compartment standard SRPX03 2012 D-E* 99.90 18.20 5.40 3.25 2.15 600 SOLAS 2009 + EC 2003/25 (water on deck), 2,5 m Hs * D-E = Double Ended Ferry, S-E = Single Ended Ferry SRPX01 31.9 M DOUBLE ENDED FERRY This is a small double ended Ropax ferry with an open Ro/Ro deck. The ship is operating in sheltered waters servicing a small Danish island. A picture of the ferry can be seen in figure 1. The ferry was built in 1999 fulfilling the deterministic SOLAS 90 1-compartment standard. It is fitted with 6 main transverse watertight bulkheads below the bulkhead deck with the engine room located amidships. Above the bulkhead deck there is a small passenger accommodation in starboard side and engine casing and stores rooms in port side. The ferry is approved for 100 persons and is only fitted with life rafts as the majority of the small Danish domestic ferries. A General Arrangement of the ferry can be seen on figure 2 on the next page. Fig. 1 Picture of SRPX01 Erichsen Small Ro/Pax Vessel Stability Study 3

Fig. 2 General Arrangement Drawing of SRPX01 Erichsen Small Ro/Pax Vessel Stability Study 4

SRPX05 49.9 M SINGLE ENDED FERRY This is a small single ended Ropax ferry with an enclosed drive through Ro/Ro deck. The ship is operating in sheltered waters servicing a Danish island. A picture of the ferry can be seen in figure 3. The ferry was built in 1999 fulfilling the deterministic SOLAS 90 1-compartment standard. It is fitted with 7 main transverse watertight bulkheads below the bulkhead deck with the engine room located amidships. At the Ro/Ro deck the ferry is fitted with side casings in each side including stair casings leading up to the passenger accommodation. The ferry has two maximum draughts assigned 2.64 m and 2.70 m. At 2.64 m draught the ferry is allowed to operate with 395 persons onboard where-as only 250 persons are allowed when the ferry is operating at 2.70 m draught. The ferry is not fitted with lifeboats. A General Arrangement of the ferry can be seen on figure 4 below. Fig. 3 Picture of SRPX05 Fig. 4 General Arrangement Drawing of SRPX05 Erichsen Small Ro/Pax Vessel Stability Study 5

SRPX04 68.5 M SINGLE ENDED FERRY This is a small single ended Ropax ferry with an enclosed drive through Ro/Ro deck. The ship is operating in sheltered waters servicing a larger Danish island. A picture of the ferry can be seen in figure 5. The ferry was built in 1997 fulfilling the deterministic SOLAS 90 2-compartment standard. It is fitted with 11 main transverse watertight bulkheads below the bulkhead deck with the engine room located amidships. At the Ro/Ro deck the ferry is fitted with side casings in each side including stair casings leading up to the passenger accommodation. The ferry is approved for 580 persons and is only fitted with life rafts as the majority of the small Danish domestic ferries. A General Arrangement of the ferry can be seen on figure 6 below. Fig. 5 Picture of SRPX04 Fig. 6 General Arrangement Drawing of SRPX04 Erichsen Small Ro/Pax Vessel Stability Study 6

SRPX06 95.8 M DOUBLE ENDED FERRY This is a larger double ended Ropax ferry with an enclosed drive through Ro/Ro deck. The ship is operating in sheltered international waters. A picture of the ferry can be seen in fig. 7. The ferry was built in 1989 fulfilling the deterministic SOLAS 74/83 2-compartment standard. It was later upgraded to fulfill the SOLAS 90 1-compartment standard. It is fitted with 11 main transverse watertight bulkheads below the bulkhead deck with the larger pump room located amidships. It has B/5 longitudinal bulkheads fitted between frame 28-64 and frame 76-112. At the Ro/Ro deck the ferry is fitted with side casings in each side including stair casings leading up to the passenger accommodation. The ferry is approved for 399 persons and is fitted with one lifeboat in each side accommodating 52 persons each. A General Arrangement of the ferry can be seen on figure 8 below. Fig. 7 Picture of SRPX06 Fig. 8 General Arrangement Drawing of SRPX04 Erichsen Small Ro/Pax Vessel Stability Study 7

SRPX03 99.9 M DOUBLE ENDED FERRY This is a double ended Ropax ferry with an enclosed drive through Ro/Ro deck. The ship is operating in sheltered waters servicing two larger Danish islands. A picture of the ferry can be seen in figure 9. The ferry was built in 2012 fulfilling the probabilistic SOLAS 2009 standard as well as the deterministic Reg. 8 side damages ( minor damages) and the bottom damage standard in Reg. 9.8. Further, the ferry is also fulfilling the water on deck deterministic damage stability standard of the EC Directive EC 2003/25 for significant wave height H s of 2.5 m. It is fitted with 12 main transverse watertight bulkheads below the bulkhead deck and watertight B/5 longitudinal bulkheads. At the Ro/Ro deck the ferry is fitted with a center casing including stair casings leading up to the passenger accommodation. The ferry is approved for 600 persons and is only fitted with life rafts as the majority of the small Danish domestic ferries. Fig. 9 Picture of SRPX03 A General Arrangement of the ferry can be seen on figure 10 below. Fig. 10 General Arrangement Drawing of SRPX03 Erichsen Small Ro/Pax Vessel Stability Study 8

RESULTS OF THE DAMAGE STABILITY STUDY All calculations have been carried out using the NAPA stability software release 2014.4-3. A lot of effort has been done in order to retrieve the original NAPA stability models developed by consultants or shipyards at the newbuilding stage. Where this has not been possible a new NAPA model has been developed and verified against the old data. Further all approved stability documentation has been retrieved and examined. In this connection, it has been ensured that unprotected and weathertight openings were taken into account. This included also examination whether progressive flooding could occur through damaged piping systems. For ships equipped with cross flooding pipes or ducts these were also modelled and taken into account as intermediate stages. Another important parameter in SOLAS 2009 is the introduction of A-class fire-rated bulkheads. Such bulkheads were also modelled and taken into account as intermediate stages. For each of the selected ships the SOLAS 2009 initial conditions were chosen based on the loading condition information in the approved Stability Booklets as well as the original damage stability GM limit curves. Further, for all the ships, it was noted that there are significant margins between the damage stability limit curve and the loading conditions from the approved Stability Booklets. In order to investigate how much index could actually be obtained a new SOLAS 2009 calculation was thereby also carried out using optimized GM limits, i.e. GM values close to the actual GM values in the different sailing conditions. SRPX01-31.9 m Double Ended Ferry Two different damage stability calculations have been carried out for this ship. The original SOLAS 90 1-compartment damage stability limit curve can be seen in blue color from figure 11 below. Based on this curve the GM values for the SOLAS 2009 damage stability calculation has been derived. These GM values are shown as GM 1990 in table 2 below next to the draughts for each of the initial conditions D s, D p and D l. The initial conditions are also plotted in figure 11 and the red line represents the SOLAS 2009 GM limit curve. The plotted points K*01 to K*08 represent the loading conditions from the approved Stability Booklet. As there is a significant margin between the plotted loading conditions and the blue and red limiting curves, an additional SOLAS 2009 damage stability calculation has been carried out based on optimized GM limits so that the loading conditions only marginally comply. This curve is shown in yellow color in figure 11 and the GM values are shown as GM opt in table 2. Table 2 Initial Conditions for the SOLAS 2009 Calculations Draught GM 1990 GM opt Deepest Subdivision D s 2.20 2.05 2.24 Partial Subdivision D p 2.07 1.72 2.28 Light Service Draught D l 1.88 1.67 2.72 Fig. 11 GM Limiting Curves for SRPX01 Erichsen Small Ro/Pax Vessel Stability Study 9

The results of the SOLAS 2009 probabilistic damage stability calculation can be seen in table 3 below. Here A 1990 is the attained subdivision index using the old SOLAS 90 GM limits and A opt is the index using the optimized GM values. Table 3 Total Attained Index SRPX01 A 1990 = 0.7542 Based on SOLAS 90 GM limits A opt = 0.7841 Based on optimized GM limits R = 0.6827 Required index fulfilled It is noted that the required subdivision index is fulfilled with a good margin. Using the optimized and thereby higher GM values is increasing the attained subdivision index from 0.7542 to 0.7841, i.e. an increase of 0.03. However, in order to achieve this, the GM limits were raised significantly, ref. table 2 on the previous page. In table 4 below, the attained index on each of the three draughts can be seen. It is noted that the ship has no problem in fulfilling the required index requirement of 0.9R on each draught. Table 4 Attained Index on each draught SRPX01 SOLAS 90 GM limits Optimized GM limits A s = 0.7538 0.7538 Passenger Ship A p = 0.7538 0.7538 0.9R = 0.6144 A l = 0.7557 0.9038 Fulfilled According to Regulation 8 of SOLAS 2009, Part B-1, Chapter II-1 the ship must be able to survive 1-compartment deterministic damages. As the ship is already a SOLAS 90 1-compartment standard ship it has no problem in fulfilling the requirement obtaining a survivability factor of at least 0.9 in all these damages. SRPX05-49.9 m Single Ended Ferry As the ferry has two maximum draughts assigned 2.64 m and 2.70 m, several calculations have been carried out. Further, this ferry has an enclosed Ro/Ro deck so the influence of the new survivability factor s covering damage to Ro/Ro decks has also been investigated. Due to the fact that the ferry is operating with different number of persons onboard at the different assigned maximum draughts, the required subdivision index will also vary slightly. At 2.64 m draught one calculation based on the original SOLAS 90 1-compartment limits has been carried out. The original SOLAS 90 1-compartment damage stability limit curve can be seen in blue color from figure 12 below. Based on this curve the GM values for the SOLAS 2009 damage stability calculation has been derived. These GM values are shown as GM 1990 in table 5 below next to the draughts for each of the initial conditions D s, D p and D l. The initial conditions are also plotted in figure 12 and the red line represents the SOLAS 2009 GM limit curve. The plotted points K*01 to K*12 represent the loading conditions from the approved Stability Booklet. Table 5 Initial Conditions for the SOLAS 2009 Calculations Draught GM 1990 Deepest Subdivision D s 2.64 2.38 Partial Subdivision D p 2.53 2.26 Light Service Draught D l 2.36 2.34 Fig. 12 GM Limiting Curves for SRPX05 395 persons Erichsen Small Ro/Pax Vessel Stability Study 10

The results of the SOLAS 2009 probabilistic damage stability calculation can be seen in table 6 below. Here A 1990 is the attained subdivision index using the old SOLAS 90 GM limits together with the original SOLAS 2009 and A 1990,SLF is the attained subdivision index using the old SOLAS 90 GM limits and the new covering damage to the Ro/Ro deck. Table 8 Initial Conditions for the SOLAS 2009 Calculations Draught GM 1990 GM opt Deepest Subdivision D s 2.70 2.19 2.25 Partial Subdivision D p 2.56 2.07 2.43 Light Service Draught D l 2.36 2.11 2.94 Table 6 Total Attained Index SRPX05 395 Persons A 1990 = 0.7577 Based on SOLAS 90 GM limits and SOLAS 2009 A 1990,SLF = 0.7173 Based on SOLAS 90 GM limits and new SLF55 ( water on deck ) R = 0.7101 Required index fulfilled It is noted that the required subdivision index is obtained with some margin. However, using the new SLF55 will significantly reduce the attained index and the required index is thereby only marginally fulfilled. In table 7 below, the attained index on each of the three draughts can be seen. It is noted that the ship has no problem in fulfilling the required index requirement of 0.9R on each draught. Table 7 Attained Index on each draught SRPX05 SOLAS 2009 New SLF55 A s = 0.7757 0.7370 Passenger Ship A p = 0.7485 0.7093 0.9R = 0.6391 A l = 0.7402 0.6942 Fulfilled According to Regulation 8 of SOLAS 2009, Part B-1, Chapter II-1 the ship must be able to survive 1-compartment deterministic damages. As the ship is already a SOLAS 90 1-compartment standard ship it has no problem in fulfilling the requirement obtaining a survivability factor of at least 0.9 in all these damages when using the SOLAS 2009. However, the regulation 8 damages are not fulfilled when using the new SLF55. The minor damages of regulation 8 are thereby the limiting factor for this ship and not the probabilistic calculation. From table 8 it should be noted that, for this ship, the GM limiting values can be raised significantly on the lower draughts but not as much on the deepest subdivision draught. The results of the SOLAS 2009 probabilistic damage stability calculation using the SOLAS 90 GM limits can be seen in table 9 below. Here A 1990 is the attained subdivision index using the old SOLAS 90 GM limits together with the original SOLAS 2009 and A 1990,SLF is the attained subdivision index using the old SOLAS 90 GM limits and the new covering damage to the Ro/Ro deck. Table 9 Total Attained Index SRPX05 250 Persons A 1990 = 0.6950 Based on SOLAS 90 GM limits and SOLAS 2009 A 1990,SLF = 0.6693 Based on SOLAS 90 GM limits and new SLF55 ( water on deck ) R = 0.6974 Required index not fulfilled It is noted that the required subdivision index is not fulfilled in the two calculations. This clearly shows the significant influence the freeboard has on the attained subdivision index calculation. In table 10 below the attained index on each of the three draughts can be seen. It is noted that the ship has no problem in fulfilling the required index requirement of 0.9R on each draught. Table 10 Attained Index on each draught SRPX05 SOLAS 2009 New SLF55 A s = 0.7004 0.6727 Passenger Ship A p = 0.7007 0.6747 0.9R = 0.6277 At 2.70 m draught the ship is allowed to carry 250 persons. For this option two different damage stability calculations have been A l = 0.6731 0.6519 Fulfilled carried out. The original SOLAS 90 1-compartment damage Apart from the significant influence from the freeboard, one of stability limit curve can be seen in blue color from figure 13 on the other factors influencing the damage stability results for this next page. Based on this curve the GM values for the SOLAS ship is the design of the engine room amidships. In starboard 2009 damage stability calculation has been derived. These GM side of the engine room the ship is fitted with a stair casing and values are shown as GM 1990 in table 8 next to the draughts for in port side a smaller emergency escape. Both of these are each of the initial conditions D s, D p and D l. The initial surrounded by A-class fire rated bulkheads. This can be seen in conditions are also plotted in figure 13 and the red line the red clouds at the General Arrangement drawing in figure 4. represents the SOLAS 2009 GM limit curve. The plotted points K*01 to K*15 represent the loading In SOLAS 90 only the watertight subdivision were taken into conditions from the approved Stability Booklet for this option. account in the damage stability calculations. However, in As there is a significant margin between the plotted loading SOLAS 2009, also the influence of A-class fire rated bulkheads conditions and the blue and red limiting curves, especially on are taken into account. The background of this is that the A- the lower draughts, an additional SOLAS 2009 damage stability class bulkhead might resist some water pressure before it calculation has been carried out based on optimized GM limits collapse so the influence of this must be taken into account. so that the loading conditions only marginally comply. This Especially the stair casing in starboard side has an influence on curve is shown in yellow color in figure 13 and the GM values the result as this will create asymmetrical damages. are shown as GM opt in table 8. Erichsen Small Ro/Pax Vessel Stability Study 11

The results of the SOLAS 2009 probabilistic damage stability calculation using the optimized GM limits can be seen in table 11 below. Here A opt is the attained subdivision index using the optimized GM limits together with the original SOLAS 2009 s- factor and A opt,slf is the attained subdivision index using the optimized GM limits and the new covering damage to the Ro/Ro deck. Table 11 Total Attained Index SRPX05 250 Persons A opt = 0.7667 Based on optimized GM limits and SOLAS 2009 A opt,slf = 0.7453 Based on optimized GM limits and new SLF55 ( water on deck ) R = 0.6974 Required index fulfilled In table 12 below, the attained subdivision index on each of the three draughts can be seen. It is noted that the ship has no problem in fulfilling the required index requirement of 0.9R on each draught. Table 12 Attained Index on each draught SRPX05 SOLAS 2009 New SLF55 A s = 0.7172 0.6887 Passenger Ship A p = 0.7753 0.7587 0.9R = 0.6277 A l = 0.8485 0.8318 Fulfilled By raising the GM limits so that the loading conditions only marginally fulfil these requirements, the required subdivision index is obtainable with some margin. However, the GM values on the lower draughts have been raised significantly. Fig. 13 GM Limiting Curves for SRPX05 250 persons Erichsen Small Ro/Pax Vessel Stability Study 12

SRPX04-68.5 m Single Ended Ferry Two different damage stability calculations have been carried out for this ship. The original SOLAS 90 2-compartment damage stability limit curve can be seen in blue color from figure 14 below. Based on this curve the GM values for the SOLAS 2009 damage stability calculation has been derived. These GM values are shown as GM 1990 in table 13 next to the draughts for each of the initial conditions D s, D p and D l. The initial conditions are also plotted in figure 14 and the red line represents the SOLAS 2009 GM limit curve. The plotted points K*00 to K*08 represent the loading conditions from the approved Stability Booklet. As there is some margin between the plotted loading conditions and the blue and red limiting curves, especially on the lower draughts, an additional SOLAS 2009 damage stability calculation has been carried out based on optimized GM limits so that the loading conditions only marginally comply. This curve is shown in yellow color in figure 14 below and the GM values are shown as GM opt in table 13. Table 13 Initial Conditions for the SOLAS 2009 Calculations Draught GM 1990 GM opt Deepest Subdivision D s 2.90 2.46 2.60 Partial Subdivision D p 2.74 2.34 2.60 Light Service Draught D l 2.49 2.36 3.15 The results of the SOLAS 2009 probabilistic damage stability calculation using the SOLAS 90 GM limits can be seen in table 14 below. Here A 1990 is the attained subdivision index using the old SOLAS 90 GM limits together with the original SOLAS 2009 and A 1990,SLF is the attained subdivision index using the old SOLAS 90 GM limits and the new covering damage to the Ro/Ro deck. Table 14 Total Attained Index SRPX04 A 1990 = 0.8940 Based on SOLAS 90 GM limits and SOLAS 2009 A 1990,SLF = 0.7873 Based on SOLAS 90 GM limits and new SLF55 ( water on deck ) R = 0.7252 Required index fulfilled It is noted that the required subdivision index is fulfilled with a huge margin in both calculations. This ship has a much higher freeboard than SRPX05 so this again clearly indicates the significant influence the freeboard has on the attained subdivision index calculation. However, it should also be noted that, using the new SLF55, will again significantly reduce the attained index. Fig. 14 GM Limiting Curves for SRPX04 Erichsen Small Ro/Pax Vessel Stability Study 13

In table 15 below, the attained index on each of the three draughts can be seen. It is noted that the ship has no problem in fulfilling the required index requirement of 0.9R on each draught. Table 15 Attained Index on each draught SRPX04 SOLAS 2009 New SLF55 A s = 0.8907 0.7983 Passenger Ship A p = 0.8943 0.7860 0.9R = 0.6527 A l = 0.9001 0.7678 Fulfilled According to Regulation 8 of SOLAS 2009, Part B-1, Chapter II-1 the ship must be able to survive 2-compartment deterministic damages as it is carrying more than 400 persons. As the ship is already a SOLAS 90 2-compartment standard ship it has no problem in fulfilling the requirement obtaining a survivability factor of at least 0.9 in all these damages when using the SOLAS 2009. When using the new SLF55 s- factor ( water on deck ) one damage is marginally below 0.9s and is thereby not fulfilling the requirements. The minor damages of regulation 8 are thereby the limiting factor for this ship and not the probabilistic calculation. The results of the SOLAS 2009 probabilistic damage stability calculation using the optimized GM limits can be seen in table 16 below. Here A opt is the attained subdivision index using the optimized GM limits together with the original SOLAS 2009 s- factor and A opt,slf is the attained subdivision index using the optimized GM limits and the new covering damage to the Ro/Ro deck. Table 16 Total Attained Index SRPX04 A opt = 0.9100 Based on optimized GM limits and SOLAS 2009 A opt,slf = 0.8038 Based on optimized GM limits and new SLF55 ( water on deck ) R = 0.7252 Required index fulfilled By raising the GM limits significantly so that the loading conditions only marginally fulfils these requirements, the attained subdivision index for this ship will thus only change marginally, i.e. from 0.8940 to 0.9100. SRPX06-95.8 m Double Ended Ferry This is the ferry that was included in the study in order to investigate how a pre-solas 90 ship would perform under the SOLAS 2009 requirements. Three different damage stability calculations have been carried out for this ship. The original SOLAS 74/83 (1983 amendments to SOLAS) 2-compartment damage stability limit curve can be seen in blue color from figure 15 on next page. Based on this curve the GM values for the SOLAS 2009 damage stability calculation has been derived. These GM values are shown as GM 74/83 in table 18 next to the draughts for each of the initial conditions D s, D p and D l. The initial conditions are also plotted in figure 15 and the red line represents the SOLAS 2009 GM limit curve. The ship was later upgraded to fulfil the SOLAS 90 1- compartment standard with a maximum of 399 persons onboard. The SOLAS 90 1-compartment damage stability limit curve can be seen in orange color in figure 15. Based on this curve the GM values for the SOLAS 2009 damage stability calculation has been derived. These GM values are shown as GM 1990 in table 18. The initial conditions are also plotted in figure 15 and the yellow line represents the SOLAS 2009 GM limit curve. The plotted points K*00 to K*09 represent the loading conditions from the latest approved Stability Booklet. As there is some margin between the plotted loading conditions and the limiting curves, especially on the lower draughts, an additional SOLAS 2009 damage stability calculation has been carried out based on optimized GM limits so that the loading conditions only marginally comply. This curve is shown in black color in figure 15 and the GM values are shown as GM opt in table 18. Table 18 Initial Conditions for the SOLAS 2009 Calculations Draught GM 74/83 GM 1990 GM opt Deepest Subdiv. D s 3.60 1.89 1.86 1.98 Partial Subdiv. D p 3.46 1.80 1.72 1.98 Light Service D l 3.26 1.77 1.64 2.73 The results of the SOLAS 2009 probabilistic damage stability calculation using the SOLAS 74/83 GM limits can be seen in table 19 below. Here A 74/83 is the attained subdivision index using the old SOLAS 74/83 GM limits together with the original SOLAS 2009 and A 74/83,SLF is the attained subdivision index using the old SOLAS 74/83 GM limits and the new s- factor covering damage to the Ro/Ro deck. In table 17 below, the attained index on each of the three draughts using the optimized GM limits can be seen. It is noted that the ship has no problem in fulfilling the required index requirement of 0.9R on each draught. Table 17 Attained Index on each draught SRPX04 SOLAS 2009 New SLF55 A s = 0.8966 0.8038 Passenger Ship A p = 0.9082 0.7989 0.9R = 0.6527 A l = 0.9403 0.8138 Fulfilled Table 19 Total Attained Index SRPX06 A 74/83 = 0.6888 Based on SOLAS 74/83 GM limits and SOLAS 2009 A 74/83,SLF = 0.6489 Based on SOLAS 74/83 GM limits and new SLF55 ( water on deck ) R = 0.7068 Required index not fulfilled It is noted that the required subdivision index is not fulfilled in the two calculations. Compared to the other ships it is noted that the GM limits are much lower. Further, the freeboard is also low for this size of ship so this again clearly indicates the significant influence the freeboard has on the attained subdivision index Again the significant influence of the new covering Ro/Ro deck damages should be noted. calculation. Erichsen Small Ro/Pax Vessel Stability Study 14

In table 20 below, the attained index on each of the three draughts can be seen. It is noted that the ship has no problem in fulfilling the required index requirement of 0.9R on each draught when the SOLAS 2009 is used. However, the requirement is not fulfilled on the deepest subdivision draught when using the new SLF55 ( water on deck ). Table 20 Attained Index on each draught SRPX06 SOLAS 2009 New SLF55 A s = 0.6733 0.6303 Passenger Ship A p = 0.6926 0.6530 0.9R = 0.6362 A l = 0.7125 0.6778 Not Fulfilled The results of the SOLAS 2009 probabilistic damage stability calculation using the SOLAS 90 GM limits can be seen in table 21 below. Here A 1990 is the attained subdivision index using the SOLAS 90 GM limits together with the original SOLAS 2009 s- factor and A 1990,SLF is the attained subdivision index using the old SOLAS 90 GM limits and the new covering damage to the Ro/Ro deck. Table 21 Total Attained Index SRPX06 A 1990 = 0.6639 Based on SOLAS 90 GM limits and SOLAS 2009 A 1990,SLF = 0.6213 Based on SOLAS 90 GM limits and new SLF55 ( water on deck ) R = 0.7068 Required index not fulfilled As the GM limits are lower than the original SOLAS 74/83 limits, the attained index is thereby also lower and it is noted that the required subdivision index is far from being fulfilled. The results of the SOLAS 2009 probabilistic damage stability calculation using the optimized GM limits can be seen in table 22 below. Here A opt is the attained subdivision index using the optimized GM limits together with the original SOLAS 2009 s- factor and A opt,slf is the attained subdivision index using the optimized GM limits and the new covering damage to the Ro/Ro deck. Table 22 Total Attained Index SRPX06 A opt = 0.7519 Based on optimized GM limits and SOLAS 2009 A opt,slf = 0.7201 Based on optimized GM limits and new SLF55 ( water on deck ) R = 0.7068 Required index fulfilled By raising the GM limits significantly so that the loading conditions only marginally fulfils these requirements it is thereby shown that the required subdivision index requirement can be fulfilled. In table 23 below, the attained index on each of the three draughts using the optimized GM limits can be seen. It is noted that the ship can now fulfil the required index requirement of 0.9R on each draught also when using the new SLF55. Table 23 Attained Index on each draught SRPX06 SOLAS 2009 New SLF55 A s = 0.6994 0.6594 Passenger Ship A p = 0.7392 0.7061 0.9R = 0.6362 A l = 0.8822 0.8696 Fulfilled Fig. 15 GM Limiting Curves for SRPX06 Erichsen Small Ro/Pax Vessel Stability Study 15

SRPX03-99.9 m Double Ended Ferry This is the newest ferry built in 2012. It is built under SOLAS 2009 and was thereby designed to comply with the SOLAS 2009 damage stability requirements. This includes also the regulation 8 minor damages and the bottom damage standard under regulation 9.8. As the ferry is a class C ship under the European Union legislation it is also complying with the water on deck deterministic damage stability standard of the EC Directive EC 2003/25 for significant wave height H s of 2.5 m. Three different damage stability calculations have been carried out for this ship. The original SOLAS 2009 damage stability limit curve can be seen in red color from figure 16 below. It should be noted that the GM limits are extremely high, i.e. between 5.86 m and 7.40 m. This is mainly caused by the intact stability GM curve which can be seen in green color on figure 16. The limiting factor resulting in the very high GM limits for intact stability is the weather criteria. The weather criteria are a part of the mandatory intact stability criteria and are given in Part A, section 2.3, of the 2008 Intact Stability Code. The criteria were not in force when the other ships included in this study were built and has thereby not been applied for those ships. For smaller ships having a high superstructure compared to the draught these criteria have a significant effect on the stability compliance. Based on this curve the GM values for the SOLAS 2009 damage stability calculation has been derived. These GM values are shown as GM 2009 in table 24 next to the draughts for each of the initial conditions D s, D p and D l. In order to compare with the SOLAS 90 2-compartment standard a new SOLAS 90 deterministic damage stability calculation was carried out. The SOLAS 90 2-compartment damage stability limit curve can be seen in dark blue color in figure 16. Based on this curve the GM values for the SOLAS 2009 damage stability calculation has been derived. These GM values are shown as GM 1990 in table 24. The initial conditions are also plotted in figure 16 and the yellow line represents the SOLAS 2009 GM limit curve. In order to compare with the water on deck deterministic damage stability standard of the EC Directive EC 2003/25 another deterministic damage stability calculation was carried out. The EC 2003/25 2-compartment damage stability limit curve can be seen in light blue color in figure 16. Based on this curve the GM values for the SOLAS 2009 damage stability calculation has been derived. These GM values are shown as GM wod in table 24. The initial conditions are also plotted in figure 16 and the dotted yellow line represents the SOLAS 2009 GM limit curve. The plotted points C01 to C12, shown in figure 16, represent the loading conditions from the latest approved Stability Booklet. Table 24 Initial Conditions for the SOLAS 2009 Calculations Draught GM 2009 GM 1990 GM wod Deepest Subdiv. D s 3.25 5.86 2.99 3.41 Partial Subdiv. D p 3.07 6.44 3.05 3.50 Light Service D l 2.80 7.40 3.28 3.82 Fig. 16 GM Limiting Curves for SRPX03 Erichsen Small Ro/Pax Vessel Stability Study 16

The results of the SOLAS 2009 probabilistic damage stability calculation using the original GM limits can be seen in table 25 below. Here A 2009 is the attained subdivision index using the original SOLAS 2009 and A 2009,SLF is the attained subdivision index using the new covering damage to the Ro/Ro deck. Table 25 Total Attained Index SRPX03 A 2009 = 0.9565 Based on original GM limits and SOLAS 2009 A 1990,SLF = 0.8851 Based on original GM limits and new SLF55 ( water on deck s- factor) R = 0.7272 Required index fulfilled It is noted that the required subdivision index is fulfilled with a huge margin in both calculations. Apart from the very high GM limits, this ship also has a much higher freeboard than the other ships, so this again clearly indicates the significant influence the freeboard has on the attained subdivision index calculation. However, it should also be noted that, using the new SLF55 s- factor, will again significantly reduce the attained index. In table 26 below, the attained index on each of the three draughts can be seen. It is noted that the ship has no problem in fulfilling the required index requirement of 0.9R on each draught. Table 26 Attained Index on each draught SRPX03 SOLAS 2009 New SLF55 A s = 0.9452 0.8858 Passenger Ship A p = 0.9633 0.8924 0.9R = 0.6544 A l = 0.9653 0.8690 Fulfilled The results of the SOLAS 2009 probabilistic damage stability calculation using the SOLAS 90 GM limits can be seen in table 27 below. Here A 1990 is the attained subdivision index using the old SOLAS 90 GM limits together with the original SOLAS 2009 and A 1990,SLF is the attained subdivision index using the old SOLAS 90 GM limits and the new covering damage to the Ro/Ro deck. Table 27 Total Attained Index SRPX03 A 1990 = 0.8651 Based on SOLAS 90 GM limits and SOLAS 2009 A 1990,SLF = 0.7947 Based on SOLAS 90 GM limits and new SLF55 ( water on deck ) R = 0.7272 Required index fulfilled Table 28 Attained Index on each draught SRPX03 SOLAS 2009 New SLF55 A s = 0.8533 0.7942 Passenger Ship A p = 0.8651 0.7941 0.9R = 0.6544 A l = 0.8888 0.7967 Fulfilled The results of the last calculation using the water on deck limit curve can be seen in table 29 below. Here A wod is the attained subdivision index using the original SOLAS 2009 s- factor and A wod,slf is the attained subdivision index using the new covering damage to the Ro/Ro deck. Table 29 Total Attained Index SRPX03 A wod = 0.9073 Based on EC2003/25 water on deck GM limits and SOLAS 2009 A wod,slf = 0.8368 Based on EC2003/25 water on deck GM limits and new SLF55 ( water on deck ) R = 0.7272 Required index fulfilled It is noted that the required subdivision index is still fulfilled with a huge margin in both calculations. However, again it should be noted that, using the new SLF55, will significantly reduce the attained index. In table 30 below, the attained index on each of the three draughts can be seen. Table 30 Attained Index on each draught SRPX03 SOLAS 2009 New SLF55 A s = 0.8911 0.8334 Passenger Ship A p = 0.9125 0.8406 0.9R = 0.6544 A l = 0.9295 0.8361 Fulfilled According to Regulation 8 of SOLAS 2009, Part B-1, Chapter II-1, the ship must be able to survive 2-compartment deterministic damages as it is carrying more than 400 persons. The ship has no problem in fulfilling the requirement obtaining a survivability factor of at least 0.9 in all these damages when using the SOLAS 2009. When, using the new SLF55 s- factor some damages does not comply when using the SOLAS 90 GM limits. This is in line with the results of the other ships. For this ship the bottom damage standard of SOLAS Chapter II- 1, Part B-2, regulation 9.8, has also been applied. It is observed that all defined bottom damages are complying with the stability requirement in regulation 9.8 based on the different GM limits studied. It is noted that the required subdivision index is still fulfilled with a huge margin in both calculations. In table 28, the attained index on each of the three draughts can be seen. It is noted that the ship still has no problem in fulfilling the required index requirement of 0.9R on each draught. Erichsen Small Ro/Pax Vessel Stability Study 17

WEATHER CRITERIA From the calculation of the newest ship SRPX03 it is noted that the weather criteria results in very high GM limits. The weather criteria are a mandatory requirement included in Part A of the 2008 Intact Stability Code. However, the criteria was not in force in Denmark when the older ships included in this study was built so it has only been requested for the newest ship SRPX03. The criteria address the ability of a ship to withstand the combined effects of beam wind and rolling. The criteria are applied as follows: 1. the ship is subjected to a steady wind pressure acting perpendicular to the ship s centreline which results in a steady wind heeling lever (lw1), 2. from the resultant angle of equilibrium (ϕ 0 ), the ship is assumed to roll owing to wave action to an angle of roll (ϕ 1 ) to windward. The angle of heel under action of steady wind (ϕ 0 ) should not exceed 16 or 80% of the angle of deck edge immersion, whichever is less, 3. the ship is then subjected to a gust wind pressure which results in a gust wind heeling lever (lw2); and 4. under these circumstances, area b shall be equal to or greater than area a, as indicated in figure 18 below. A possible reduction of the wave induced moment would thereby be much more relevant for the small passenger ships which are operating in sheltered Danish waters. The effect of applying the weather criteria has been analysed for the ships SRPX05, SRPX04 and SRPX06. Applying the criteria will increase the intact stability GM limits by approx. 1-2 m and none of the vessels will be able to operate based on these requirements. This is shown by the limit curves in figure 19 on the next page. The dotted green curves represent the original intact stability limit curves without weather criteria compliance and the blue curves are based on weather criteria compliance. In order to compare the effect of applying the weather criteria with the SOLAS 2009 attained subdivision index, new damage stability calculations has been carried out. In practise GM values derived from the weather criteria GM limit curves as shown in figure 19 (blue colour curves) have been used as a basis for the calculations. The total attained indices based on the SOLAS 2009 can be seen from table 31 below: A 1990 A weather SRPX05 = 0.6950 0.8527 SRPX04 = 0.8940 0.9250 SRPX06 = 0.6639 0.8264 Table 31 Attained Index using the weather criteria GM limits It is seen that the attained indices will be ranging between 0.8264 and 0.9250 for the different ships when using the GM limits derived from the weather criteria. This is a surprisingly large increase. Fig. 18 Severe Wind and Rolling Criteria where the angles in figure 18 are defined as follows: ϕ 0 = angle of heel under action of steady wind, ϕ 1 = angle of roll to windward due to wave action, ϕ 2 = angle of down-flooding (ϕ f ) or 50 or ϕ c, whichever is less, ϕ f = angle of heel at which openings in the hull, superstructures or deckhouses which cannot be closed weathertight immerse. Φ c = angle of second intercept between wind heeling lever lw2 and GZ curves. The wind heeling levers are based on a standard wind pressure of 504 Pa corresponding to a wind velocity of 26 m/s with uniform velocity profile. The value of wind velocity used for ships in restricted services may be reduced to the satisfaction of the Administration. However, for ships operating in Danish waters they might face the standard wind velocity but not the wave induced moment. Erichsen Small Ro/Pax Vessel Stability Study 18

Fig 19 Influence of the weather criteria on SRPX05, SRPX04 and SRPX06 Erichsen Small Ro/Pax Vessel Stability Study 19

CONCLUSION Even though the study shows mixed results, and not the anticipated very high attained subdivision indices for all of the calculated ships, some very important conclusions can be drawn. First of all, it is to be underlined that, for the studied ships, it is the minor damages of SOLAS 2009 regulation 8 which is the limiting factor and thereby not the required subdivision index. Secondly, the freeboard has a very significant influence on the attained subdivision index results. The ships having a high freeboard obtain the highest attained subdivision indices. This can also clearly be seen from figure 17 below where the attained and required indices for each ship are plotted against the freeboard divided by the moulded breadth. Thirdly, it is observed that all of the ships will comply with the required subdivision index if the GM limits are raised so that the loading conditions in the approved Stability Booklets only marginally complies with these curves. This is due to the fact that all of the ships, especially on the lower draughts, have a margin to the limiting curves. A design parameter which can have a very significant influence on the SOLAS 2009 damage stability results is the introduction of A-class fire rated bulkheads in the damage stability calculations. The design of passenger ships must be carefully considered when it comes to the location of A-class fire rated bulkheads as asymmetrical designs will cause asymmetrical damages which can have a significant negative effect on the damage stability results. The introduction of the new SLF55 for the damages involving the Ro/Ro deck is also having a significant effect on the attained index calculations as well as on the minor damages according to regulation 8. From the ships included in this study it is shown that the reduction in attained index due to the introduction of the new SLF55 is between 3% and 12% with an average of 7%. This is much higher than the 1-3% reduction seen for larger Ropax ships in other studies [1]. It is also shown that the weather criteria included in Part A of the 2008 Intact Stability Code results in very high intact stability GM limits for these small ferries. Most of the small Ropax ships are designed for small ports and thereby limited in their main dimensions. Further, the draught is often also limited as the ships are operating in sheltered waters. Increasing the draught in order to fulfil the weather criteria might thereby not be a solution for many designs. A solution could be to reduce the superstructure and thereby also the freeboard of the ship. However, that will have a significant negative effect on the damage stability compliance as shown in this study. If the small Ropax ships should fulfil the weather criteria in its present form one of the only options are thereby increased GM limits. REFERENCES Germany+CESA: Improving the survivability of ro-ro passenger ships, SLF55/INF.13, 14 December 2012. Skoupas, Sotiris: Investigation of the Impact of the Amended s- factor formulation on Ropax ship, STAB 2015, June 2015. Fig. 17 Attained and Required Subdivision Index as a function of freeboard and breadth Erichsen Small Ro/Pax Vessel Stability Study 20