^-. USADACS Technical Library inn um A ' // "/ DECISION RISK ANALYSIS FOR DYNAMIC SEPARATORS. "\s. NORMAN H. TRIER THOMAS N.

Transcription

1 F^-ADM.*"^ RA-76-U423 ;.' ^-. USADACS Technical Library inn um v AD V "«-^TT \ f - >\ ' i ' i y />> '.4; x v : \ r;*,a, /h V? v A ' // "/ DECSON RSK ANALYSS FOR DYNAMC SEPARATORS > >^v A)- : r -S S-v w j i > * *. > '\ > > i/' j _ X *, y v M V' "\s. NORMAN H. TRER THOMAS N. MAZZA JAMES B. BEESON 1 '< 1. 'S -> 4> ; ' K 1 r>" ^!,/ -, 'Y f : t>j f 1 ' MAY 1976,-1 '") 4-'-"/^ -A'< A ) ") - L Approved for public release; distribution unlimited. <>; 4? ;.1: '" --;, ; v -. t.; vf/ N " -J ) ^" ' US ARMY ARMAMENT COMMAND >("' \ '. X-i SYSTEMS ANALYSS DRECTORATE ROCK SLAND, LLNOS V-- M. - ' <; (:>V4> r ')'

2 ' * 1 \ * ' * "s ' >? v «\ ^; -4. -, *. -.> i- -. ^. DSPOSTON ' Destroy this report when no longer needed. Do not return it to the originator. 1 -v DSCLAMER ~V v o V.. -' j The findings in this report are not to be construed as an official Department of the Army position. >. ' '..- ' -., -v-. WARNNG nformation and date contained in this document are : phased on input available at the time of preparation. ( Because the results may be subject to change, this document should not be construed to represent the official position of the Army Materiel Development and Readiness Command ualep&v so stated v A -' >:.."'. hr'l s, ^ - > '. k / s. ; V <. v "- N. i-, \, y - A : \ / r '. > ' v " ) V / V ' ; ' r" ' V inj." ( * s- V} -> } ; '\.. 1 K' 'V ( * 1

3 UNCLASSFED SECURTY CLASSFCATON OF THS PAGE (When Data Entered) READ NSTRUCTONS REPORT DOCUMENTATON PAGE BEFORE COMPLETNG FORM 1 REPORT NUMBER 2. GOVT ACCESSON NO 3. RECPENT'S CATALOG NUMBER DRSAR/SA/N TTLE (and Subtitle) DECSON RSK ANALYSS FOR DYNAMC SEPARATORS 5. TYPE OF REPORT & PEROD COVERED Note - Final 6. PERFORMNG ORG. REPORT NUMBER 7. AUTHORf«) Norman H. Trier Thomas N. Mazza James B. Beeson 9. PERFORMNG ORGANZATON NAME AND ADDRESS US Army Armament Command Systems Analysis Directorate (DRSAR-SA) Rock sland, L CONTROLLNG OFFCE NAME AND ADDRESS US Army Armament Command Systems Analysis Directorate (DRSAR-SA) Rock sland, L CONTRACT OR GRANT NUMBER*"«) 10. PROGRAM ELEMENT, PROJECT. TASK AREA & WORK UNT NUMBERS 12. REPORT DATE May 191$ 13. NUMBER OF PAGES MONTORNG AGENCY NAME & AODRESSf*/ different from Controlling Office) 15. SECURTY CLASS, (of rhie report) UNCLASSFED 15a. DECLASSFCATON DOWNGRADNG SCHEDULE 16. DSTRBUTON STATEMENT (of this Report) Approved for public release; distribution unlimited. 17. DSTRBUTON STATEMENT (of the abgtract entered n Block 20, f different from Report) 18. SUPPLEMENTARY NOTES 19. KEY WORDS (Continue on reverse aide t nee e a aery and identify by block number) Dynamic Separators Risk TNT Manufacturing Static Separators Quantity-Distance (0D) 0. ABSTRACT (Continue on reverse aide t neceeamry and dentify by block number) nvestigations resulting from the explosion of the TNT facilities at Radford AAP concluded that excessive amounts of explosive material were contained within the Nitration and Purification (N&P) Building and Quantity Distance (OD) requirements would be violated if the lines were reconstructed in their former configuration. One method of satisfying QD requirements is to increase the acid to nitrobody ratio in the N&P section. To maintain the present production rate, flow rates would have to be increased beyond the capability of the current static separators. t has been proposed DD/, FORM AN EDTON OF 1 NOV 65 S OBSOLETE UNCLASSFED SECURTY CLASSFCATON OF THS PAGE (When Data Entered)

4 UNCLASSFED SECURTY CLASSFCATON OF THS PAGE(Wi«n Datm Entmtmd) that dynamic separators be purchased and installed in place of the static separators to accommodate the required increased flow rate. Dynamic separators are expected to achieve the required increase in-process flow without major structural changes. This analysis examines the risk and the expected time and cost of installing dynamic separators at Radford with and without a prior Manufacturing, Methods, and Technology (MM&T) project. t was concluded that the MM&T would not reduce the risk and would delay the reconstruction unnecessarily. #ii UNCLASSFED SECURTY CLASSFCATON OF THS PkGE(Wh*n Data Entered;

5 CONTENTS OBJECTVE 5 Page NTRODUCTON 5 ALTERNATVES 6 APPROACH 7 RESULTS 7 CONCLUSONS 15 APPENDX A. OTHER CONSDERATONS 17 APPENDX B. NETWORK DESCRPTON WTH TME AND COST NPUT DATA 21 APPENDX C. TERMNAL NODE HSTOGRAMS FOR EACH ALTERNATVE 27 DSTRBUTON LST 87 Next page is blank.

6

7 OBJECTVE The objective of this analysis was to determine if any benefits could result from conducting a Manufacturing, Methods, and Technology (MM&T) project at the Volunteer Army Ammunition Plant (VAAP) or Joliet Army Ammunition Plant (JAAP) prior to installing dynamic separators during the reconstruction of B and C TNT lines at Radford Array Ammunition Plant (RAAP). NTRODUCTON When facilities which contain explosive material are constructed, the distance required between neighboring structures depends upon the quantity of detonable material contained within each structure. When TNT lines A, B, and C were constructed at Radford AAP, the distance between structures was calculated on 12,000 pounds of class 7 (detonable) material. On 31 May 1974, TNT line A at Radford AAP was destroyed by an explosion. Lines B and C were also damaged, which initiated an investigation to determine how much explosive material was actually contained in line A at the time of the explosion. The investigation concluded that there was 18,000-19,000 pounds of class 7 material being processed in the Nitration and Purification (N&P) building at the time of the explosion, not counting remelt TNT which is sometimes present in this building. This was sufficiently above the original design quantity to declare lines B and C unsafe in their present configuration. A restoration project for B and C lines was submitted as a late-start 76 project. The project would repair B and C lines in their present location. To meet the new Quantity Distance (QD) requirements and maintain a production rate of 50 tons per day, either structural or process changes must be implemented. A structural solution to this problem would be to build a retaining wall between the Nitration and Purification Section in the TNT building. However, examination of the TNT facilities and the structural requirements of a retaining wall showed this to be a high-risk alternative. Thirty foot of separation would be required between the Nitration and Purification Sections to allow time for a detonation trap to activate in the event an explosion occurred. This would require major structural and process changes. This alternative was not pursued due to the high expectations of the process changes discussed in this study. The process change under consideration would reduce the explosive quantity to the original design specification by increasing the acid-tonitrobody ratio in the nitration process from 3:1 to 9:1 and increase the flow rate accordingly to maintain the production rate of 50 tons per day. n order to accommodate the required increased flow rate, dynamic separators would be installed in place of the present static separators. The gravitational action of the current static separators limits the possible flow rate; however, by utilizing centrifugal force, dynamic separators are expected to achieve the required increase in the process flow without major structural modifications. The proposed flow rate requires

8 dynamic separators to process 30,000 liters/hour in the nitration section and 10,000 liters/hour in the purification section. AB Bofors of Sweden, a developer of dynamic separators, has considerable experience in the construction of 10,000 liters/hour separators for TNT lines and 10,000 and 20,000 liters/hour separators for nitroglycerine lines. They have also constructed three 20,000 liters/hour separators for the Canadian ndustries Limited (CL) lines. The CL units will be installed on a 35 ton per day TNT production line in September AB Bofors has not produced any 30,000 liters/hour units. They have indicated size can be misleading and there would be some risk in scaling up to 30,000 liters/ hour, since an increase in capacity will not necessarily result in a commensurate increase in output. The risks associated with the operation of a TNT line with dynamic separators are categorized as mechanical or process flow problems. Mechanical problems may arise due to sludge build-up in the separator or uncontrolled vibrations of the rotating basket. Process flow problems may arise with the increased flow rate. f the retention time in the nitrators does not permit sufficient nitration, the flow rate would have to be reduced. This would increase the TNT equivalence in the N&P building and QD requirements would not be met. ALTERNATVES After reviewing the project and outlining the necessary steps required to rebuild lines B and C, four primary alternatives were identified for evaluation. The primary difference among the four alternatives is the time when approval is given to purchase the required dynamic separators. The alternatives are as follows: ALT 1. mmediate approval is given to purchase and install 28 dynamic separators (14 for each TNT line). After the TNT lines have been rebuilt with dynamic separators, one line would be proven-out to determine if QD requirements are met. ALT 2. mmediate approval is given to purchase and install 14 dynamic separators on one TNT line B, while static separators are installed on TNT line C. f operation with dynamic separators is successful and QD requirements are met, then dynamic separators would be purchased and retrofitted on TNT line C. ALT 3. Approval to purchase 28 dynamic separators is withheld until three of them are tested at VAAP or JAAP. This proposed MM&T project would replace two static separators in the Nitration Section and one in the Purification Section with dynamic separators of various sizes. The TNT line would be operated, using current operating procedures (e.g., flow rates) to correct any mechanical problems which may arise. Therefore, this would not demonstrate concept feasibility (i.e., whether a TNT line can operate at the proposed higher flow rates and if this will lower the TNT equivalence enough so QD requirements are met).

9 ALT 4. Approval to purchase dynamic separators is withheld until Canadian ndustries Limited (CL) installs and operates their dynamic separators. After the CL units are evaluated, a decision would be made to purchase and install dynamic separators on Radford AAP's two TNT lines (ALT 1) or proceed with further testing (ALT 3). As in ALT 3, the CL installation would operate at current flow rates and not demonstrate concept feasibility. Delivery of the dynamic separators could be expected approximately six months after the order is placed, and it will take approximately two years to rebuild lines B and C. As preliminary testing could be performed at JAAP or VAAP while waiting on the construction, ALT 1 was broken into two parts: ALT la. mmediate approval is given to purchase and install the dynamic separators, but no preliminary testing would be performed. ALT lb. mmediate approval is given to purchase and install the dynamic separators, but preliminary testing is performed at JAAP or VAAP while waiting on the construction at Radford AAP. APPROACH Each alternative course-of-action was represented by a network of activities from the initial decision to success or failure of that alternative. For example, the decision to adopt ALT 2 initiates activities to rebuild line C with static separators and rebuild line B with dynamic separators. These are represented as lines (arcs) A5 and A6 of the network in Figure 1. Follow-on activities are represented through final prove-out (A52) when the course-of-action will be judged a success or failure, depending on whether the QD requirements are met. The condition in which the separators function properly but QD requirements are not fully achieved has been labeled conditional success. Minimum, maximum, and most likely time and cost estimates were obtained for each activity depicted in Figure 1; see Appendix B, Table Bl. Probabilities associated with the possible outcomes of tests and likely decisions were also estimated and are presented in Figure 1 (e.g., it was estimated that the probability of failing preliminary testing at VAAP (A30) is.04 (A31). These data were combined (using the Venture Evaluation and Review Technique (VERT)) in a computer simulation to generate time/cost estimates with uncertainty for each alternative. RESULTS Time and cost values for each activity were accumulated from the time a course of action was chosen to the time when the results of operating a TNT line with dynamic separators were evaluated. These values include Next page is blank.

10

11 ALT 2 A5 Rebuild line w7stättc^ep A6 Rebuild line w7a'yn^sep"! N D T E R M M C A4 A19 A34 ALT 4 A N D ALT 1 A39 Evaluate A41.85 A40 A42 A15 Signal Arc ALT la A20.5 Construction & nstall- ation of dyn. sep w/o test A21.5 Decide to perform pre^mttestsa? VAAP or JAAP ALT lb A22 Prove-Out A30 Pr«Un,. Teatin«at VAAP or JAAP A29 A23 k.9 Success "A24 "".06 Conditional success 1 A N D»4C failure A N D E R M Retrofit both lines w/static sep. E R M A N D T E R M A53 Success.9 A54 A.08 C onditional Success N M A D C.02 N D A N D T E R M T E R M A55 Failure 0 R Retrofit both lines w/static separators E R M CnnKl-rnrMon w/n installation of separators Dynamic Separators A43 ALT 3 A44 MM&T Project A4 6.9 Success AA5.1 Failure Construction and nstallation of Dynamic separators A N D A L L A47 Prove-Out.9 A48 Success A N D A49.08 Conditional Success JB. E R M A51 Failure Retrc7Tt"T>öTnT3nes"' w/static separators A N D E R M FGURE 1. Dynamic Separator Network Analysis

12

13 the time and cost of retrofitting static separators if needed, but do not include the time and cost for further structural or process modifications which would be required if QD requirements were not met. The expected time and cost of each possible outcome for each alternative, with 90% probability intervals (ninety percent of the values were contained between the upper (95%) and lower (5%) values), are presented in Table 1. This table summarizes the terminal node histograms from the VERT computer simulations which are located in APPENDX C. For ALT la there is a 90% chance that operation with dynamic separators would meet QD requirements. The expected time to complete the project is 31 months at an expected cost of $8,73M. The 90% probability interval (ninety percent of the values were contained between the upper (95%) and lower (5%) values) for time and cost are 29 to 32 months and $8.10M to $9.26M, respectively. There is a 10% chance that operation with dynamic separators would not meet QD requirements and additional process and/or structural changes would be required. This consists of a 6% chance that the dynamic separators would remain on the TNT line and a 4% chance that static separators would be retrofitted. The outcomes for ALT lb are similar to ALT la, except that the expected cost for successful completion of the project is increased by $0.17M. This increase is due to the preliminary testing at JAAP or VAAP. Even with preliminary testing, the chance remains at 90% that operation with dynamic separators would meet QD requirements. However, because of preliminary testing, the probability is reduced from 4% to 2% that static separators would have to be retrofitted because operation with dynamic separators is infeasible. For ALT 2 there is a 90% chance that operation with dynamic separators would meet QD requirement. f this occurs, dynamic separators would be purchased and installed on the second TNT line at Radford AAP. The expected time to complete this project is 40 months at an expected cost of $8.73M. The 90% probability intervals for time and cost are 37 to 42 months and $8.29M to $9.07M. ALT 2 requires 9 months more than ALT 1 if QD requirements are met; however, there is a 10% chance that the QD requirements would not be met. n the latter case, the purchase and installation cost ($0.75M) of dynamic separators for the second TNT line would not be incurred. For ALT 3, if dynamic separators are installed after the completion of the tests, the expected^ time to complete the project is 71 months at an expected cost of $8.92M. The 90% probability intervals for time and cost are 66 to 76 months and $8.26M to $9.48M. A substantial increase in time is incurred since a MM&T project would not start until 1976 and would require approximately 18 months (and $0.2M) to complete. There is a 10% chance that upon completion of the MM&T, dynamic separators would not be purchased and static separators would be installed. f this occurred, the time to complete_the project would be months at an expected cost of $7.44M + $0.5M. 11

14 TABLE 1. COST/SCHEDULE COMPARSON WTH 90% CONFDENCE NTERVAL Schedule (Mo) Cost ($M) Decision 5% Expected 95% 5% Expected 95% ALT la. Rebuild lines B and C w/dynamic separators w/o MM&T or any preliminary testing ! Possible Outcomes 1. Operation w/dynamic separators meets QD requirements. 2. Operation w/dynamic separators does not meet QD requirements. 3. Operation w/dynamic separators is not possible. Static separators are installed ALT lb. Rebuild lines B and C w/dynamic separators w/o MM&T, but w/preliminary testing Possible Outcomes 1. Operation w/dynamic separators meets QD requirements. 2. Operation w/dynamic separators does not meet QD requirements. 3. Operation w/dynamic separators is not possible. Static separators. are installed

15 TABLE 1 (Cont'd) Schedule (Mo) Cost ($M) Decision 5% Expected 95% 5% Expected 95% ALT 2. nitially rebuild one line w/dynamic separators and the other w/static separators. No MM&T Possible Outcomes 1. Dynamic separators completely successful and installed on both lines. 2. Operation w/dynamic separators does not meet QD requirements. Dynamic separators installed on one line'only Operation w/dynamic separators is not possible. Static separators installed on both lines ALT 3. Perform MM&T project at JAAP or VAAP before rebuilding B and C lines Possible Outcomes 1. nitially rebuild w/dynamic separators. a. Operation w/dynamic separators meets QD requirements. b. Operation w/dynamic separators does not meet QD requirements. c. Operation w/dynamic separators is not possible. Static separators are installed nitially rebuild w/static separators

16 TABLE 1 (Cont'd) Schedule (Mo) Cost ($M) Decision 5% Expected 95% 5% Expected 95% ALT 4. Defer decision until Canadian dynamic separators are evaluated Possible Outcomes 1. Rebuild w/dynamic separators. No MM&T. a. Operation w/dynamic separators meets QD requirements. b. Operation w/dynamic separators does not meet QD requirements. c. Operation w/dynamic separators is not possible. nstall static separators Rebuild w/dynamic separators after MM&T. a. Operation w/dynamic separators meets QD requirements. b. Operation w/dynamic separators does not meet QD requirements. c. Operation w/dynamic separators is not possible. nstall static separators Rebuild w/static separators only

17 For ALT 4, if the decision were made to purchase and install dynamic separators at Radford after the Canadian units were evaluated, the expected time to complete the project is 42 months at an expected cost of $8.74M. The 9j0% probability intervals for this outcome are 39 to 44 months and $8.09M to $9.26M. An increase of 11 months is incurred since the Canadian units would not be evaluated until late CY75 or early CY76. Then, if it were decided to perform a MM&T project before purchasing and installing dynamic separators at Radford AAP, an additional 18 to 20 months would be needed. CONCLUSONS Preliminary testing of the dynamic separators would correct mechanical problems which would arise and reduce the probability from 4% to 2% that static separators would have to be retrofitted because operation with dynamic separators was infeasible. The probability that operation with dynamic separators will meet QD requirements is not increased by performing any preliminary testing, since successful operation is based on operating the TNT line at the increased flow rates. Because of this, there is little to be gained by ALT lb, ALT 3, and ALT 4. Rebuilding both lines with dynamic separators (ALT la) has a lower expected completion time than ALT 2 (initially proceeding with dynamic separators on only one TNT line); however, ALT 2 concludes with the same information on dynamic separators as ALT la, but without the additional sunk cost for dynamic separators. f operation with dynamic separators does not meet QD requirements, the expected cost of ALT 2 is $8.0M in comparison to $8.76M for ALT la. With ALT 2, time is sacrificed in converting the second line if dynamic separators are successful; however, there does not appear to be any reasons to require an earlier completion date based on current mobilization requirements as outlined in APPENDX A. 15 Next page is blank.

18

19 APPENDX A ADDTONAL CONSDERATONS Next page is blank. 17

20

21 According to the Production Base Analysis (PBA for FY76-802, tne Level mobilization requirements for TNT production is 47.64AM lbs/mo. This TNT requirement increases to 66M lbs/mo if CMs can not be produced in the quantities stated in the PBA and HE rounds must be substituted in their place. The TNT lines which are designated to produce the required TNT are shown in Table A-l. Of the 22 lines listed in Table A-l, two of the TNT lines at Joliet are the manual batch lines; all others are continuous. TABLE A-l. TNT MOBLZATON REQUREMENTS Location No. of Lines Capacity (M ///Mo) Newport 5 15 Volunteer 6 18 Joliet 8 24 Canadian Manufacturing 1 2 Radford a 2 6 Total 65M ///Mo. Assuming B and C lines are rebuilt. The number of lines that are currently capable of producing TNT are listed in Table A-2. TABLE A-2. PRESENT TNT PRODUCTON CAPABLTY Location No. of Lines CL Type Batch Production Capability (M ///Mo) Newport Volunteer Joliet a Canadian Manufacturing Radford 0 a Joliet has 5 additional batch lines which could produce TNT if supplied with enough oleum. To obtain this extra oleum, a new plant could be built at an estimated cost of $24M. 19

22 As shown in Table A-2, without Radford's production, there is still enough TNT production capability to meet mobilization requirements. This would require the use of additional batch lines in place of the continuous lines. However, the present planning (Table A-l) includes the use of two batch lines and Joliet indicates that the batch lines can be activated with less risk in meeting the mobilization schedule. Therefore, the question concerning the actual need for B and C lines at Radford should be readdressed, based on an economic criteria. There is still some question as to the actual TNT equivalence in the N and P building (estimates range from 12,000-20,000 pounds). t has been proposed that Picatinny Arsenal perform a test to determine the actual TNT equivalence in the N and P buildings of continuous TNT lines. Five months after Picatinny Arsenal is funded, the tests should be completed. f the results indicate that the TNT equivalence is less than 12,000 pounds, then operation of the TNT lines with static separators at Radford would meet QD requirements. f Radford is being used as a vehicle to prove out dynamic separators, then a counterargument can be made that it would be cheaper to install them at an already existing line, such as at Joliet, Volunteer, or Newport Army Ammunition Plants. 20

23 APPENDX B NETWORK DESCRPTON WTH TME AND COST NPUT DATA Next page is blank. 21

24

25 TABLE B-l. NETWORK DESCRPTON Name ACTVTY Description TME (Mo) COST ($5) A4 Decide to restore line B with dynamic separators and line C with static separators. A5 Restore line C with static separators A6 Restore line B with dynamic separators A7, A8 A15, A21 A34, A60 A61 Signal Arc. A9, A22 A47, A52 Prove-out the TNT line which has the dynamic separators A10 TNT line prove-out was unsuccessful. Static separators are necessary All, A23 TNT line prove-out was successful. Operation with dynamic A48, A53 separators meet QD requirements. A12, A24 TNT prove-out determined that dynamic separators operate A49, A54 mechanically, but QD requirements are not met. A17 Given that operation with dynamic separators meets QD requirements, then install dynamic separators on the other TNT line A19 Decide to restore both B and C lines with dynamic separators. A20, A46 Restore B and C lines with dynamic separators (Projects and ). Cost + 10% deviation A29 Same as A20 above, except that the time and cost for installation of half of the dynamic separators has been subtracted

26 TABLE B-l.(Cont'd) ACTVTY TME COST Name Description (Mo) ($M) A30 While restoration of B and C lines is taking place, install a dynamic separator at JAAP or VAAP to test its mechanical ability. Leadtime, installation, testing, removal, and return shipment to Radford are considered. A31 Failure of larger size dynamic separators to function properly. Static separators are necessary. A32 Dynamic separators will work mechanically, but major modifications are needed which would delay completion of project. Renegotiation of contracts, changes to equipment, and leadtime are considered. A33 Dynamic separators work properly with minor modifications. No slip in program. A62 f dynamic separators are rejected as a result of the preliminary testing at VAAP or JAAP, then B and C lines will be completed with static separators. A37 f dyanmic separators are excepted as a result of the preliminary testing at VAAP or JAAP, then install them on B and C lines at RAAP. A38 Defer the decision to rebuild E and C lines until the Canadian dynamic 9-9-li separators are evaluated. A39 Evaluate the Canadian dynamic separators A40 Rebuild B and C lines w/static separators only A41 Based on favorable Canadian results, rebuild B and C lines with dynamic separators. A42 Based on questionable Canadian results (their 20,000 liters/hr units work, but it's possible that the proposed 30,000 liters/hr units will not work) do a MM&T at VAAP or JAAP.

27 TABLE B-l.(Cont'd) ACTVTY TME COST Name Description (Mo) ($M) A43 Decide to do a MM&T project at VAAP or JAAP before B and C lines are restored. Money could be funded as a late-start FY76 or a FY77 project. A44 A45 MM&T project; includes negotiation, leadtime, installation and testing. Based on unfavorable results from the MM&T, rebuild B and C lines with static separators A51, A26, TNT Prove-out was unsuccessful. Static separators are necessary A ,25 N3 2! rn - Do ft

28

29 APPENDX C TERMNAL NODE HSTOGRAMS FOR EACH ALTERNATVE Page PART 1. TME AND COST DSTRBUTONS FOR ALT la 29 PART 2, TME AND COST DSTRBUTONS FOR ALT lb 39 PART 3. TME AND COST DSTRBUTONS FOR ALT 2 51 PART 4. TME AND COST DSTRBUTONS FOR ALT 3 59 PART 5. TME AND COST DSTRBUTONS FOR ALT Next page s blank.

30

31 PART 1. TME AND COST DSTRBUTONS FOR ALT la Next page is blank. 29

32

33 ALT. R H.«S67 29*lt> * *? * l.6«U *99 32,5693 3? «'J69 34.U62 3*.4*5* S.04* ? ' 36.9U10 HO, OflS. = loo«me/ «OD PART 1. TME * : Composite Tiae Diatribution U MN * *» ** 51» * * «0,039 * 33 * MAX STD ERROR PEARSONAN SKE* AND COST DSTRBUTONS FOR AT la CPU MN 28'. «567 * »« ! »» 0* ««« «««««««««««««« »»»...» ^ «««««««« Z ^ ««««««« ««««««««««««««««««««««««««««««««««««««««««0# «« ««««* *5693»«««««««««««««««» «««««««««««« » #» *» *» *» * «o, ^ «««« ! o ! «««.gar !** ** * Ü * » ««««««« « ««o »»»»»»» * « o,999 36i»820 ««. ««««« /» * * *»» * * i, / ' MAX COEF OF VARATON «4 KURTOSS (BETA 2) * 5,

34 g NO. nhs. PATH AND OVERALL COST ARE THE SAME FOR THE COMPOS!Tt TERMNAL NUDE ALT.1A: Composite Cost Distribution RFO , MN 7.»U2 T 7.V7S * !* «««8.2331! *. 975! « r { « ! j «««8.6838! «8.748? ! 8.9*13 1!** ! «««9.1632! ! *.4563 * 9, <J.S«51 9.S851 loflo MEAN * MODE , , , , * MAX STD ERROR * 8,7701 PEARSONAN SKEW CFO 0, * ,8 0* MN 7.9H * ! ! »««*«« ! * ««0.1* ! «««« » « » #»» *.»»« » »#* ««««« «««« T69 i.; :.*,» ««. * * *»» « «««# t*tt..«t«t»*t«f»*«.t»«*»t»t»»*.tft«ft«« »*«.»***«»»»««*» « ««..*»*»*.» ^»»».».«*»» «.*. 0* * «* «««* o « *.»««* *»««o*** * » «*».»..» *» *» « A563»» «««««« ««« MAX COEF OF VARATON 0.O4 KURTOSS (BETA 2) v -^^r-,;. -ti^ *?.' s af* *=. ;%.:*, - m. - v j0&;- ^% *3>J:"

35 w NETWORK TME FOR NODE N42 ALT.1A: Operation s Successful and QD Requireaents Are Met. RFD U.05 O.lü 0, CFO ,6 0, ,9 1,0 2b.» HN 28, MN 0,0 28, Q.1Ö41 # 29.3*77 «29.t>114! ] 29.«388! 3025 ;» ! ! # ! ] 30.« ! ; ! 31.»032 »« ! !* 32,2943!* 32.4U80 : ! 3?,7854 3?.» NO. OPS. m 9)3 MEAN * HODE» , , ! !» f 29,6751! ! 3025!» ! # 30,4936»». «*»»* *.. 30,6873! 30,8210 ^». ».,. 30, ,1484 *«* «*«* #«««31*3121 -;'» «* «* 31,4758 * * 31,6395! «» * * *» *« J! ««#*»» «* ««< »» i , ,032 0,047 0,068 oat« t *»»» ««0, ,024 *»#««. «* # * ia HAK MAX STO ERROR COEF OF VARATON 0,03 KURTOSS «BETA 2) «2, PEARSONAN SKEW a 0.4Q *7 0,

36 _ w PATH AND OVERALL COST ARE THE SAME FOR MODE N42 ALT.lA; Operation s Successful and QD Requirenents Ars Met. WFO U »H2 1- «1 1 *N ,0 7,*112 7, f ! ,1687»»». 32 8, ! # ! ,4263!«* 61 « ( «S ! * 64 B ! * d3» «*« T ,1988 0,030 9, ,3276 0,018 9, * »851 MAX B6 STD ERROR * 0,3 *.»125 ] «* 8.#769 {# ** ! ** 9»*W! «.0701! 9,i344! »632! ! r.4563! $851 MO. ftps. * 913 MEAN = MOOE * 8,7749 PEARSONAN SKEW ,1 0,2 0, *5 0,6 0, *-! 1 1, 1 1 «x 1 1 * *» es * «*»»»» ««» «« « «a*»*»»** «««» » ««ft******* **»» *» ** * * *» ««*» **» * ** ««# # * * * 0,977 «1.0 MN ,959 0,982 0,997 1,008 0,0 MAX 21 COEF OF VARATON 0,04 KURT0SS (BETA 2) «2,45

37 NO. nps. «NfcTwOKK TME FOR NOOE M43 ALT.1A; Operation s Successful. But QD Sequin its Are Not Me* RFÜ CFO Ü.3 0,* ,8 0* * HN * MN * * *5 ** *5 20, «0.11* *} ** 32** *5 i»» * *04 *»««**»«* 0.11* b i *1!* * * C.*Ö72 3Ö c.7186!* 30.»3* *! * 31.6** ! ! 3? ?.3383! 33.*5*0 1 *«32.S ** MEAN * MODE «*5 68 *5 68 f.» 23 * * * *5 MAX TO STO ERROR * PEARSONAN SKEW 30.*872» *««* ««» » ««30.83*3 t * * «* * «" * * *» * *# « "-!»» «* ««*»«««* «« «««««««««31.*128 *»»*<»»»*«« * **» * **«««***»**»***»*»*»**« *» * «««»»«*» «* #»*»» « ».»» * «««««** #»» * < S »W !» < «««o »*5*0 » o, *»» *»» ««»* ««i g « MAX COEF OF VARATON 3 KURT0SS (BETA 2) * V-v - >r --&. -

38 PATH ANÜ OVERALL COST ARE THE SAME FOR NOOE N43 *_T.1A: Operation s Successful, But QD Requirements Are Not Met RFO Q.Q ft HN B.0519»*** *«ft.2528 ft.3030»«** ! **»*»»*«* ** ff» *» 8.7*4? »* E570 9, NO. OBS. «44 MEAN» * CFD , , MN ! ! ! !* ! « **»«**«*«* ,3532» # 0, ! *» ! «0, ! « ! *» 0, ««* * ^» ««« « * *» # *» { * » ««*»» *" 45 *» «« o Mtl*flMM4HM4MMtf«tMMMMMM«MM« ' 23 *»««««««««0, ,023 ««««««* «««« «* * «0, » ««9, * « «*» «*«* «*»» , »»» ** *» *«0, ! * ««« !» ^ «« «* , * «**o» « * « «^ MAX 0.«9,3575 MAX STD ERROR» 0,3328 COEF OF VARATON 0,04 KURTOSS (BETA 2) * 2,47 MODE PEARSON1AN SKEW 0.24 S&aÄ?-.

39 ALT.1A: NETKORK THE FtfR NODE N*4 Static Separators Are nstalled And QD Requirements Are Not Met RFD 5 0.1«ü.l«j 'J007-1 X - MN 3?.0007! *91! 23 3?.3776 J } 47 3?.5661 l 70 3?,7545! 23 3?,9430! * * ! * ,5084 i l »! «853 34,0737 ]* * 93 " 34,1622 l» «* T **** 0, «76 1 r «* r ! ! * 0, »814!# «# H , l MAX NO. nfls. «43 MEAN « STü ERROR A HU>T*ODAL DSTRBUTON CFO , * l -f 1 MN ! 0,023 32,1891! ! 0, X*«.«*»»«# t«32,9430! « ,1314! « !»» * * * ,5684!» «>»* « , ! ««« ^»« «« t«mt<wio*mt>mm»m>nm*»*m ! ** * ««0, ! *««* * « *»*«««« «** *»» «« ^» ««» a ^ ««««a X ** + * * +*+ + * + o * ** ** * * * * o *7699» * «#«*.»» aa*»aaa»*a«* o.fss l***:^»»w^^»*^»*»»^* ++** * ^». » ^. 9*m*+V ^ «^ * *»» <>»»»»» *» ***»»! , MAX COEF OF VARATON 3 KURTOSS (BETA 2) 2.33

40 , ' '" ' g PATH AND OVERALL COST ARE THE SAME FOR NOOE N44 ALT.XA; Static Separators Are nstalled And QD Requirements Are Not Met HFO »- --i ---- j Mira 1 o.o ! 23 ' «.1706 «# !* ] 47 «.3373 «* *! 23 «.3040»« (»# ! 93».7262 * * »»* 70 8.*3T *«23 «. 485! * » #»» , ! ! ** 0, ' ! !* * 9.6*86 47»« *2 x « MAX MO. rbs. 43 MEAN * STD ERROR * MODE» PEARSOWAN CFO 0.J * U MN * X«* ! ! ! * «**** * «« »«***»««« ! « l ~ ! *»« ««# » *»«* * ** ! «««* « !» «0*512 0«f*85! > ««« *«««»»«»» »» ««** «» *»»«« #*» «««« *»«««*»««« **. «* «. « ^ ««««» «««o.«««»«« »«« ^» « «« #»»» * # * # MAX COEF OF VARATON 4 KÜRT0SS (BETA 2) * 2.27

41 PART 2. TME AND COST DSTRBUTONS FOR ALT lb Next page is blank. 39

42

43 PART 2. TME AND COST DSTRBUTONS FOR ALT B ALT. 8 ; Composite Tiae Distribution CFO * MN !- 1-1 MN ' 27.4* * ! ! ** ! * !* ,084! «« * # # # # » » A «* #» * * «««*»» ««*«> * o *T * * 0.Ü » * ««» »* »t*<»M«#»»# <»> #»##*»»»» - ** «: -4jfjlff'- 3? <* * 38 * * o.929 3? «* * * * #*# *.«* #»» «» ?.9* # *** «*«««. «» «. « * * «« o «*» « * *»» *»» # * *»«0^9;Tl *009 1-«.»...«..» « X»«#»»«a..»» *«««.»*» «««*«« j * *» ««««**» * *»» «o # * # * #.»...*.* *««*»* *«. o »«#» i» *» ««.»» «i, t»25 MAX MAX NO. ORS. lnofi ME N = STD ERROR * W00E PEARSONAN SKEW 3 COEF OF VARU7ON «4 KURTOSS (BETA 2) 7.51 ^^%^;i*.->;.»4$^

44 ALT. Hl 7.7* * S 7.« «,1«2« fl.30o3 e.3«8l " * ^40 9, U * NO. OPS. «lf»0o *E MODE PATH 4N0 OVtRALL COST ARE THE SAME FM ti : Composite Cost Distribution D Ü MN»«#»» * ««* * « #* «* ( MAX STD trror * 9.03<!0 PEARSONAN SKEW THE COMPOSTE TERMNAL NODE CFD , , X - MN 7* X » » « ! ! ! ! « *# ! « «««««,.-;y^**x, « X «««*«««««*«*«**»***««*«**»«««* «*««**««««**».*»»**»**«««««**«*»» *. 6» X» « X ^ «* » ««««« «**»««««<»*»t*i«f«)t*lft«*h»t»*mm*«lt * * «o, » *» * «««o » * ««« * *««««! MAX COEf OF VARATON «4 KURTOSS (SETA 2)» 2«79 = 0.38

45 NETwOrtK TME FOR NODE N45 ALT.ö: Opera tion s Successful and QD Requir RFÜ O 0.25 CF ? «H,»»V4 j-» j "!«" "! WN « MN * » *** t l f+m 0.109! « B > 0.123! « * >» * ! « ! « ««<» * 0.534» 0.687» A i <» f «43! c * 20! » o.953 3? <» «0, <» o, « <» «*!» ! ! «> X « o ! «> # «« «> «> = * # ««* * # *»» * o » «« ^ ««««« «i.qoq b MAX MAX NO. 0«S.» 856 MCAN « STO trror « COEF OF VARATON» 0*04 KURTOSS (BETA 2) «11.32 MODE PEARSONAN SKEW «0,13

46 fi.1196 b.1196 fi.1865 «.2535 S «.5681 «.7220 H.7B90 B.8S59 P o.lvofi Q^726l P «599 *»««# # NO. obs. «856 «EAN «MODE *AfM ANÜ OVERALL COST ARE THE SAME FOR NODE N45 ALT,18: Operation s Successful and QD Requirements Are Met. RFD 5 O.lü j x MTN * * ,019 0.Ü , MAX STD ERROR « PEARSON1AN SKEW CFD Ü MN ! *» ! ! ! «o.l » !» o » ! « X »*»««*»****«*«****»*** ^»» !* ! ! *»»» * ! ««« o.911 9,3914 o t>922 ^ »» # 0, *» * i, ooo MAX 0,3345 COEF OF VARATON * 4 KURTOSS (BETA 2) *7

47 NO. «PS. NETWORK TME FOR NODE N*6 ALT.JP; Operation s Successful. But QD Requirements Are Not Met RFO 5 0.1U CF[ 29.5* MN ! 26 29, n.H2? 30.3Ü26 1 ** 30.**29 X * ! # t 3l.0t> ), ! * 3? ?.3961 «3?.5b64! ! 33.3* * *.2993 * * MEAN» MODE * MAA STD ERROR PEARSONAN SKEW « ! **« « *» « « * Q »»*«*» MN «« ^» « «« ^»» «« «*» o» * «« ^««« ! «««« ,1090! ««« ^ ««i.ooo MAX COEF OF VARATON «3 KÜRTOSS (BETA Z) «6.35

48 NO, nbs. PATH änü OVtRALL COST ARE THE SAME FOk NODE N46 ALT*1#J Operation s Successful, But QD Requirements Are Not Met HN P M.1899 X* 8.2*67 *««.3075 X» ! a.42so i M.5426 y.feui* i « « » *79 «« ** X«9.6242! ! O *93 1» MEAN «MQOi * 0, , * MAX STD ERROR * PEARSONAN SKEW CFO * ** » « X «****»****»* ! ! * » * «** * ««« b * COEF OF VARATON = 4 KURTOSS (BETA 2) MN , ,455 0,545 0,597 0, MAX ' 2.56

49 NO. "BS. «NETWOWN rihe FOR NOOE N47 ALT.1H: Static Separators Are nstalled And QD Requirements Are Not Met HFD CFO U MN , ! 27.45« * «* « , » « h ^ » *EAN rtooe O ,0 0.O15. MAX STD ERROR «27.68*7 PEARSONAN SKE« * * * « ^ » e » «*» «31, ** ! » #» MN 0, , O.TOi , , ,701 0, ^ «« «e , * ^ «e 0, » !»» ,0 36.*072 MAX COEF OF VARATON a 0.10 KURTOSS (BETA 2) 2,04 0,78

50 >ATH AND OVERALL COST *RE THE SAME FUR NODE N4 7 ALT. 18: Static Separators Are nstalled And QD Requirements Are Kot Met RFD 5 0.1« CFO ? MJN 7.73*0 ' 7.»107 7.««74»«*«* 7.«64! A.19* H.3476 H.4243 fr »»**»** ! 6,7312 * K.8079 «* « r* 8.«613 i 9, *7 *» i 1» 9.*b82 T 9.3**9 : i « i i <J.!*750 i* i i* i i i NO. 0B5.» 67 MEAN A MüLTMODAL DSTRBUTON C , U MAX STD ERROR « » * «« *» ,1942 8, , » *««** * ! « «*»» * * *« COEF OF VARATON * 5 KURTOSS «BETA 2) MN , MAX * 3.06

51 35.9*94 3b ft.83* S ? ?.4« ALT.?: « *.82*9 NO. ribs. * «96 ME MOOE * NfcTWOWK TME FQH Operation s Successful and QD Requirements D j MN , MAX 40,039* STD ERROR «40,7766 PEARSONAN SKEW NODE N39 Are Met. CFO MN , , «0, ,3571! ! 0, ! ! ««* > <>. o,212 38,7149 ***«*»»»»*« «««« ! »* #» * «#» *# ! ! ! ! ! ! « *» «*»* » #* *» o ! ! o **»»» «..»».» ^ b» «« , # MAX COEF OF VARATON 4 KURTOSS (BETA 2)

52 NO. oes. PATH AND OVfcHALL COST ARE THE SAME FOR A_T.2 : Operation s Successful and QD Requirements u !«-.- i 1 j N 8.095«ft.1*47 8.1* H b « , ,5363 « A.6832 B U A.S300 «.«79«* »258 9.» Q.X * 9.3ö84 «96 ME N MODE MAX B.7277 STD ERROR» REARSONAN SKE* NODE N39 Are Met. CFO MN , « ! ! ! ! *«*«**« ! * ! ! , » ! 0, «« « ! ! ! « ! « ,2705 »«9,999 9.J195! ««« * »«* * * «*»»* *»»1, MAX COEF OF VARATON * 3 KURTOSS (BETA 2) * 2.65 s 9

53 PART 3. TME AND COST DSTRBUTONS FOR ALT 2 51 Next page is blank,

54

55 S PART 3. TME AND ALT.2 : Composite Time Distribution XFO 5 Q S 2P.7396! MN 0.U 2* '29.3* /69 * ! # * J1.6J2Ö ! 34.3Ü74 34,* ,lb34 * »»* *. 193 T T *99 ** 4?.9686! , COST DSTRBUTONS FOR ALT CFO MN 44, » 1, c* MAX MAX NO. OPS. * 1000 MfAN STO ERROR MODE PEARSONAN SKEW COEF OF VARATON - 7 KURTOSS (BETA 2) « # 05 o.on , » 0.946» 0.973»» 0,995» 0.996

56 NO. OBS MEAN MODE PA n ANO OVERALL CO&T ARE T*E SAME FOR ALT.2; Composite Coat Distribution HFD % *» * * MAX 8.6S14 ST0 ERROR B.6562 PEARSONAN SKEW THE COMPOSTt TERMNAL NODE CFO U MN * * * »» »» ! « ! ! ! ! X * ^ !» * !»»»« ! !» » «o " 0, lt ooo MAX COEF OF VARATON - 4 KURTOSS (BETA 2) * *02 "Ä

57 NETWORK TME FOR NODE N40 ALT«? : Operation s Successful, But QD Requirements Are Not Met RFO U.Ob CFO U MN MN O.w 2A ! 33! * 16! *049 * 16! 29.1Ö * 0,016! * 16! ! 33! # ! 49! # 0, ! 16! » 33 l»*»ee#e#»e» S !# *-»» 33 1 « ? 49» » 16 X # ! 49 # ! * , ! ««98 ^ « !# 49 MfltHHHtHHMWtMMMtHkfMMNM !» «0,049 HMMMMtNtmtNHMtfNMtltNNH« !# 49 ^» e ! 33 ^ e , ,8600 j o,0 ^ e 0,885 3? ; ««* 66 ^» ? o.o ^» e J?.JU Q57 je 16 i»»ee»»»e»ee»»»eeeee»eeeee»»#»»e»»e»ee« !» 33 ^ «^.OOO MAX MAX NO. *BS. 61 MEAN STO ERROR COEF OF VARATON» 3 KURTOSS (BETA 2) * 2.68 MOOE * 31,0598 PEARSONUN SKEW * 0.24

58 WO. OPS. * PAM AHO OVtKAU. COST ARE THE SAME FOR NOOE N40 ALT.? : Operation s Successful, But QD Requirements Are Not Met RFO Ü.Ob O.iy * MN 0.Ü ! r«7.51j1 i! 7.t>*90 1! # ] ** ! ]» *» 7.6S66» 7.69?4! # o42!## 7.SO00! t»»**»«7. 87 S! * 7.V076 1 * # ******* ft.0152!!»««8,6511 1! * !* ** * «.1228! ft.1587!* * 8.1*46! «R.2304!»» : i i i *.338l i i i! T i M MEAN * , *9 0.Ü CFO » * »#* ! 7.Ö359 *»« ^794 « Q98 MAX 8.* STO ERROR COEF OF VARATON «3 KURTOSS (BETA 2) MODE « PEARSONAN SKEW «0.16» MN MAX * 2.06

59 NETWORK TME FOR NODE N4l ALT,8 : Static Separators Are nstalled And QD Requirements Are Not Met [ 1 -, 1 M1N 31.ÄJ81 re* ] 3?.2175 u* 0,023 32,4073 ]»* S970 ] «* 23 32,7*67 ] * ] « ^ ] ] * : : 34.1U8 u«r «««« ] ] 0, ] e» ] 35.2^32 ] [ ] 3S.6J2b ] ] b r o« b MAX NO. r>es. * 4S MEi \N * STD trror * MO( >E * J*,1690 PEARSONlAN CFD U ! 32,0278 « ! » #»«##* * »«« * «« » ««« «* « * ««*« «« » « *» « * MN M» 0.488» 0.605» «es«0.767! «o.8i4 > 8.81*»» 0.907» * 0.930» * itftm#mkflmt««tmt«ff«ttt 0.977» !»* * * ««o <J0l8»» * * «o, * «* « *»***»*»»*»«*»***** #* i.o MAX COEF OF VARATON «3 KURTOSS (BETA 2) * 4,22

60 en OO PAlU AND OVERALL COST ARE Tut SAMt FUK NOOE N*l ALT.2-* Static Separators Are nstalled And QD Requirements Are Not Met RFO * *--*- 1 1 MN 7.525* *« ! 7.6* * *« : 7.9* !* ** ft.1006»» ft , ft H.33C6 T x*»*«ft.3690 ft.* ** * NO. <>es. «*3 MEAN «MODE « *7 * *7 70 *7 23 *7 70 * *7 0,023 *7 MAX 8.05** STO ERROR PEARSONlAN SKEW CFD ? * * f MN 7.525*! «X»» * «* *7 7.6*05! X»«* X! * ! 0.1* *«*»»*» ! !» X! * 0.3*9 7.9*72! « ! «o.* !» 0# *«* * 0, o ! * ««« ^ « «««0.81* *»*» # 0.88* «««* «0.88* ** «* *»»» * «*» *073! ** **56 ^» «e *8*0 1««« ««» ««i MAX COEF OF VARATON 3 KURTOSS (BETA 2) « : Ä-'

61 Part 4. TME AND COST DSTRBUTONS FOR ALT 3 59 Next page is blank.

62

63 61.83*9 6].«3*9 '63.5/ « «69.Ü / ?.2*46 72.» S S **9* Q ALT.3: OPS. * 1000 ** *0DE PART 4. Conposit«Tine Distribution » MN o.u , *1 0, ,011 0, MAX 70.7*49 STD trror «70,4143 PEAftSONAN SKE* «TME AND COST DSTRBUTONS FOR ALT 3 CF0 0,1 0,2 0, ,9 1* * HN 61.83* * 11 64,0656 * 24 64,8091»»» 0,059 65,5527! * 0, ! o# } * « * «*» o « ««««* 0,672 72,24*6!» ,9881 ^ «««« ^ «»» * « * ,2188 ^ 0,9i 2 75,9623!»* * ««0,946 76,7058! «« **9* X *»» «« « * ««* ««<), ,6800 «««««» o,996 80,*236»» 1, , MAX 3.*82«COEf OF VARATON «5 KÜRTOSS (BETA 2)»

64 PATH A*D ovtfcall COST AR THE SAME FOR THE COMPOSTt TERMNAL NUDE ALT.3: Composite Cost Distribution KFU 5 0.1U U MN ' «MOO * of>6 07 7,28*7 * 13 7.A028 * » * b * 0, *«29 S 8.3*7*! 31 B.*655! * ! *»» « ! # *97! * fl.9^78 {» « ! !»» »»** *6 9.*101 { *2 P.5281! * * * 9.8Ö2* MAX NO. OPS. «looo MEAN « STO ERROR «MODE « PEARSONAN SKEW CFO * 0.5 0* * MN 6.812* * *86! * *7 ** *3 7.*028! ! * ! ! ! ! ! * ** ! «o.l54 8.3*7*! « *655!» 0.25* ! o.3* ! «* 0.* * *» ^ «* ! o » ^ ««0,9X3 9.*101 #««« ^«««««««««« » o«0,9v 2 9.6* * * 9.882* MAX COEF OF VARATON «6 KURTOSS (BETA 2)» *.6* * 0.19

65 ! 64.»870! ** ,7513 7*.425=» ! *! 76.i486! * ,4706 J! r 6« J.1675 NO. nbs. * «17 *4 AN «*ODE «NETWORK TME FOR NODE N48 ALT«3: Operation s Successful and QD Requirements Are Met.»FO Ü CFO ? HN { 1 MlN O.u * O.OH! ! »»« ,042! # ! ,065 ** ! # X ** « ,084 ^ #e , ,0 70 * # * * # * * ^ e * * * ^» #»»« ^ # ,029 ^»»»# # ! # # «* C.007 ^ e ^»» ,006 ^ o ^» «o, # MAX MAX STD ERROR» COEF OF VARATON «5 KURTOSS (BETA 2) « PEARSONAN SKEW» 0.*3

66 NO. nbs. a PATH ANO OVfcKALL COST ARE THfc SAME FOR NODE N48 ALT.3 : Operation s Successful and QD Requirements Are Met HN O.u «, o76 B.Ü17 fi S97 8, a.» S * *1 9.0^82 9.0* *3 9.3* *04 9.6V MEAN MODE ,012 * ** 35 ** 42» * * 55 «84» U.080»» * 0,066 ** * * 62 «0,069 * 0, * 51 0,032 0,022 0,023 0,027 0,028 0,012 0,006 0,0 MAX STO trror * CFO * % « : !* * *» * ** PEARSONAN SKEW " * *» * * »***»*»»»» MN , , ,5404! * * ,604b» «**««*»» * ,6685 *» * * «« MAX COEF OF VARATON» 4 KURTOSS (BETA 2) «2.53

67 NO. f>rs. ALT. 3 :»FÜ 6».937* - 6*.9«74 * V »«2* * ! 6A.3Ü37 T« ! !* 69.72*9 T *«7r* ! ! * * 7? ? J2 *» » b * ; !# i«7* MEAN» MODE «NCTWOHK TME FOR NODE N49 Operation s Successful, But QD Requirements Are Not Met «K«15 0,0 60 0, * CFD » «65,9349 **«« HtMHMH« * » * * « * » *» «« » * « « MAX STD ERROR « COEF OF VARATON * 5 KURTOSS (BETA 2) PEARSONAN SKEW» 2» MN MAX * 1.94

68 NO. PATH AND OVERALL COM A«E THE SAME FOR NODE N49 ALT.3 : Operation s Successful, But QD Requirements Are Not Met RFD CFD S MN O.ü » 8.2y R.3U S.4b3? ^ * *5 3.63* b42. 9,oo<»i : «.05*0 O.1039 :» * * * : *» : : ^ * ,*532 -J.4532 ftbs.» 67 ME/ *N * )E * , * « ^»»»*« *»» * «*«*««*«**»***««« ^» «««* « { « " «« MAX STO ERROR « COEF OF VARATON 0,03 KURTOSS (BETA 2) FEARSONAN SKEW * MN * MAX * 2.70

69 MO. OPS.» ALT. 6R ? H.*J8* 6n.*b75 bw.2?6< 7-J.1U8 7*.534') 70.»531 M.3'? ?. 10* 7?«629b 7j.0*« «d * »824 7<>.*015 7ft.« ^ «f) 7H H.»lt>2 r-*.vi62 1* *E MO NETMOHK TME FOR NODE N50 Static Separators Are nstalled And QD Requirements Are Not Met « MN N * E» 71 0, CFD MN ! * ! * ! « »*»» «*»« ! « *0»« ! [ ! ! * ** **** * x ** »»» * #» ! * , *" *** * ** ** ***** * ** 0,643 ^ « b 1********* * * ** * * **** *** *** »»»»» * *» 0, , * **** *** * ** ***** *** ** ** * * 0, ** * ******** ** * ** ***** * **** * G.143 1»»»»»» #»»»»» 1, MAX MAX STO fcrror « COEF OF VARATON * 4 KURTOSS (BETA 2) 3* *2 PEARSONAN SKErf * 0.39

70 S PA1«AND OVERALL COST ARE THE SAME ALT,3 : Static Separators Are nstalled And QD RFD U *> ? HN ft * ft ül «2üo 71 e.«7l2 P S , * O.S , MAX STD ERROR = NO. OBS. «14 M A HUlTlMODeL DSTRBUTON FOk NODE N50 Requirements Are Not Met CFD Ü MN ' : * o.o7i *» ! #o»# *«»** ! [» * *» ! * « [ ** ! « ! *» «0,500 9,2251»*» * » «» « » * *»« ^» ««» «0, * * *»«* !» !» *»* 0, » «* * ^» * ,7307 «»» 0,929 9,7812! «*» *» »»» * «i MAX COEF OF VARATON * 4 KURTOSS (BETA 2) «2.13

71 NO. OBS. PATii AND OVERALL COST ARE THE SÄHE FOR NODE "Si ALT.3: Static Separators Are nstalled and QD Requirements Are Not Met RFO 5 0.1U CFD 0.1 «1 1 HN A * « # ! * * ««« *»*» ! 7.1* *0! « * a02!** ! * * * ! » ,0050 * H HEAN «*6bE « O HAX STD ERROR = PEARSONAN SKEW ! ! »» » «7.6234! * * COEF OF VARATON KURTOSS (BETA 2) HN > HAX «2.23

72 NETwOKK TME FOH NODE N51 ALT. 3: Static Separators Arc nstalled And QD Requirements Are Not Met HFO (» , *- mm - -" M1N a», E -»" -» 61.83* ?.4* * 20 6^.6654 ««*« b j***««»« T 69 66,7163 ** *6*»»*»«*»« *»**»* * ! * 39 6Q.7o71 T / «« » 20 7? * * T 77.69*4 1 MAX NO. nbs. * 102 MEAN * T0 trror * «ODE s 67, 7078 PEARSONAN CFD 0, MN ! ! ! ! 0# i ! « »««««»«.«*««* o ! « !»»« «* * !»» *. * !» ^ «« ^» «* ! *» » «« !»»» ! { ««0, ««« ^ ^ » ««« ^ ««« « » ! *» U MAX COEF OF VARATON 5 KURTOSS (8ETA

73 PART 5. TME AND COST DSTRBUTONS FOR ALT A 71 Next page is blank.

74

75 PART 5. TME AND COST DSTRBUTONS FOR ALT 4 Composite Time Distribution ALT.*: *FD b Jf.i'K: "* --- i *~~" ""l" """ "N 0.^ T * 17 3fl.7o J5 # ?.0* ^ *71 * 11 4*, «H S » * 6978 o.o 6J «02 t>*.021* 01 6*.6* » j» **9 J ** d6l ** ] Ö H.SU *22 7Q.7-J2? MAX NO. ^HS.» l'itto *EAN * STD twror «M0-) * PEAHSONAN SKEW CFO « ! < ! < ! < # < < ! « < « « < 6978 < »«*«« < , MN » «0.479» 0.815» 0.880» o.891»» 0.892» ««0.892»» 0.892» o.892» it«0.892 * 0,892» A 0.892» 0.892» «ff o.892»» o.892» 0.892» 0.893» 0.900» *9» o ! <» o » ! <» 0, !» <» «!, MAX COEF OF VARATON * 0.21 KURTOSS (BETA 2)

76 ALT. 6,6076 'ft » ,7* *54 =t.2t>37 ft Ü S.7J o.3^63 9.4<» * NO. nrs.» looo ME MODE PATH AND OVtRALL COST ARE THE SAME FOR : Composite Cost Distribution D HN *»» *** * o.ooe 04 0, MAX STO ERROR «8.78«! PEARSONAN SKEW THE COMPOSTfc TERMNAL NODE CFD , MN * « ! 27 7,3174! ! ! ! 0, ! ! ! « ! ! 0, ! ««*»»«*««««*««*»*««» o * * «o «0, « »»» »»»* * «o ^«««» 0( ««««**««« » «*.»#! MAX COEF Or VARATON 6 KURTOSS (BETA 2)

77 NETWORK TME FOR NODE N42 ALT.*: Operation s Successful and QD Requirements Are Met. RFD 5 0, ,25 CFO , «N O.u ,» B X» 13» ! 14! 39.» 'j. 7863! » ** * J ; * 4? ?.? j«««««43,«« »453 « ! 44.4U J «45.7* NO. nps, * 763 *EAN ««ODE «0, ,047 0,066 0,064 0,090 0,097 0,100 9«71 0,090 0,05«0,052 1«18 0,010 0, MAX STD ERROR * PEARSONAN SKEW ! ! « ! 41.06*3 41, " , , ^» « J» * ! ««? t 73«:x :^r '.»,««* » « '! «43,2891 f--.'» «43,5672 1,0 - MN 0,0 01 0,001 0, ,033 0,066 0, ,288 0, , JMM ^» «* o, X ^ «* *1234»» * * Q ,4014»» «««0, » * o, !, MAX 1,1277 COEF OF VARATON 0,03 KURTOSS (BETA 2) * 2* S..-V-.---<««'! '' '^?$k s^^ f '""-'-A ~

78 PATH AND OVERALL COST ARE THE SAME FOR ALT.*: Operation s Successful and QD Requirements RFC HN O.w T.H9S ! 20 ^.0906 * *58 1* 1* ^.2210! 39 H.dUb? *»* 0,039! * 0,050 M.M66»*«*» 51 **** 50 '1.5*70! * T : * *»«*»* 69 fl.7425 <*!»*»» ! o.oöl j«««««s6 8,9381»**»»«0,06* «.0033! *3 9.0btt5 ««««50 Q.1337! »«* 30 9.<:o*i i! 18 Q.3293! *5 « * 12 o.s** ,5901 MAX NO. f»rs. «763 HEAN STO ERROR «A MU»TiNOPoL DSTRBUTON NODE N42 Are Met. CFO , MN » *!, ! ! «*««« ««««««« ! « J ««*««««««««« »» X ««««« « ! *25 0, * ^ «* ! » ! * ! « ! 0, *1 0.9* ! «0, i»**»*»»»»»»»»»»»»»»»»»»»»»»»»»»»»»»»»»»»»»»»»»1, MAX COEF OF VARATON «4 KURTOSS (BETA 2) 2.54

79 NETWORK TME FOR NODE N43 Operation s Successful» But QD Requirements Are Hot Met ALT.*! «FO U JW.'tfc' "!" "" j» * 20 4';.1134 T 59 *0.ilB8 * 20 4i,b<i4l 4r.7295 i 59 4G.934B * » J.7r> * ?.1670!* 39 4?.3723 ] 39 4?.5777 j ** # !# 39 4?.9b64 J#«* O.C * [ * j» * 0, U «.0419 MAX NO. r>ps. 51 «EAN» A MU.TMOD^L DSTRBUTON STD cbror «CFD ? « ! ! & »»« * » « »*********»« ««*»«****«*« # , Ml* »»» « !» *» * «*»»«0, !«« * ««!, , MAX COEF OF VARATON * 3 KURTOSS (BETA 2) 3.13

80 PATH AND OVcRAU. COST AWE THE SAME FO«NODE N43 ALT. Operation s Successful, But QD Requirements Are Not Met H ,25 6.0*5* T 1 MN fl.t><»54 3V H * ,0 rt.2159 H f».3t*b 0,039 P..3tib3 n.*31 59 *> «,e4uy H.1977 H,Vb* ÜOÖ C Q,1« b «.5227 NO. PPS. = 51 ME A MULTMODAL ÜiST»*«tiTO»< Ü MAX STO ERROR * CFD MN 0, ! ! ! ! ' ! ! » *»»«*»«« ! ! ! «« { «««« »»**«««*. * » «* ! «« ^ «««« {» * «««»««* « « «^» * * «««» o *»» «o J522 ^ « * »»..»..., ll »227 MAX COEF OF VARATON «4 KURTOSS (BETA 2) 2.24

81 MO. o«s. ALT.* ; HFO 4?.74B8 4?.7*88!* 4?.919* 4^.0900!* !* *3.*J13! !* *» 3.9*31! *>5Ü *»!* ! !* ! ** 45.34*1 *«4^.4787! *« ** : 4A.W2! l*m l Ä ! l 47.18S2 34 HCAN MODE NETWORK TME FOR NODE N44 Static Separators Are nstalled And QD Requireaents Are Not Met M1N U CFO MN! ! * ! ! « ! ! * ! !» X «* * 0# « # Q ! »«». *# ! ! » ««««* ,029»» * « « ! * * «««# # #««* ««*»* * * «0, j * *«<> «««** *» «*»» ot 97} *» *37»» ** MAX *».«***«**»»»*«*..»! MAX STO ERROR» COEF OF VARU7OM 3 KÜRTOSS (BETA 2) « PEARSONAN SKEW * 0.37

82 PAfn AND OVERALL COST ARE THE SAME FOR NOOE N44 ALT,4: Static Separators Are nstalled and QD Requirements Are Not Met SFO U *~ i --«i-- ~" MN T ft.1200.! ! ft.2391! 29 1 rt.29ö6 1!» 0.U29 «.358? 1 U.O 6.*we r i *.*7 7.3 i *.5i69 i T 0.Ü * 59 H.5964 i! * # ! U ms»» 0,029 ft.7751! * 29 H !* o 29 ft.894?! «*««««88 ft.9538 T!*«« ? 7! 59 9,1324 0, r!» SJS! *»»«« { *« ! ! 29»* T ! Z MAX NO. OBS. 14 MEAN m STO ERROR MODE * PEARbONlA CFO ! l , , ,? " ««*«««*»e»» ^ «««a »»» * V 0.5-9, ,2515 o MN ,4302»««*» ,4897 * * 0, «« MAX * COEF OF VARATON «4 KURTOSS (BETA 2) «2.72

83 NO. OßS. s ALT.4: Operation s RFO * *60 T!* ! 6«.063*! * * J Qfa8 * * *2 71.« j»e«7? ?4**l64 f_ ! «! 7S.8825 * * 76.»598 l m *« !* T # 79.79?2 T 79, MEAN «*OD * NETWOKK TME FOR Successful and QD Requirements MN MAX STD ERROR PEARSONAN SKEW NOOE *4tt Are Met. CFO MN * # ! !* *»»» ! ! « * * « o { ««* ! «» j»»««*» » ««««* ^» «« J** ««»»»»»»««« ^ «» e ^» ««««««*» <> « S 1» «#* o » «*««* * *» ««*«« ! * * * * * «« « ! ««* «« ^ »» *»»»» * *»» «« ««««««*»» * ««* »»»* * # »*»'**» "» * MAX COEF OF VARATON» 4 KURTOSS (BETA 2) 3*00 «0.47

84 w N3 HATH AND OVERALL COST ARE THE SAME FOH NOOE N48 ALT.*: Operation s Successful and QD Requirements Are Met. «PO S O 0.25 ft, MN i i a.2284! r«h.2ö91 T - *«8.3*97 * »*«0.470S" *»* « * U4 «.7740 * 1,8346! * ! ****** 8,9558 #«*»» *.1J77 ** **»# ! *«9.4*07 *»* 9.SO* *»** » NO, n«s. «91 MEAN * A MUtTfMOO*L DSTRBUTON o.ou MAX 8.94*9 STD ERROR «CFD , * ! » * ««8.3* **«* b31b * *0 ««* ** *»»***»*.» ««* V Ü »* * * *» «*** * «*» «« ! HtiiiiMiiiiminwiMti » * t> COEF OF VARATON «4 KÜRTOSS (BETA 2) MN » 1,000 MAX «2.19

85 ALT. H ? *67.3bOS « o,18öl * l,?646 7> H M NO. obs. * 10 ME N MOOE NETWORK TME FOR NODE N49 Operation s Successful, ButQD Requirements Are Not Met ,25 1! MN O.u ! ! #«eee 0#loo ! ! # ««««»«*«0, »* * o.ioo!» ^ » ,0 { « A ! ## * *# ** *» ««««* «*<>«»«» » *» ««* * «« *«* * «* * «««* «««*»» ^««««* ! # «*»» ,100 ^ « « ] « *»«» «««««*» «0, ! »» «* !* «*»«0» *»» * ** *» «««* #! MAX MAX 71.3*72 STO ERROR COEF OF VARATON «4 KURTOSS (8ETA 2) PEA«SONlAN SKEW «0.55 *«««#»**»*

86 NO. OBS. «ALT. R M.1809 ^.2264 «.271b ft « ? S «166 «. 621 H.9Ö75 8.9b?y 8.»983 9, , S MEAN MODE HAlrt AMD OVtRALL COST ARE THE SAME FOR NOOE N49 Operation la Successful, But QD Requirements Are Not Met C MN «* » «0,100 0,0 0,0 0, ,0 0, , ,0 MAX STD ERROR « PEARSONlAN SKE* CFD 0,1 0.? 0, ,5 0,6 0,7 0,8 0, MN { « ! ,2716! ! « ! ! { » ,5442! * ! ! »»» ! ! ** f !» »»«**»*»«**»« ! ! *37! {» *« *5! ! ! T07! «0, , , MAX 0,3730 COEF OF VARATON 0,04 KURTOSS (BETA 2) « v#?&l

87 ALT.* : RFD B *u *.«538 *6.1*2b *7.*Ji3 *A ) ?.5t» *0 56.4* * ,607^ 62.89fot> 6*.lbS* ^ « * NO. obs. 50 MEAN MODE NETWORK TME FOR NODE N51 Static Separators Are nstalled and QD Requirements Are Not Met O ! MN , ,4099 0, » , *««* « * »» * * * * *0 « ««it««6*.185* 65.*7* S * MN ««1,000 MAX MAX STO ERROR COEF OF VARATON 0.23 KURTOSS (BETA 2) *1 PEARSONAN SKE«« * *** » ^ ^ g'*

88 NO. BS. * HAlh AND OVERALL COST ARE THE SAME FOR NODE «51 ALT.«: Static Separators Are nstalled And QD Requirments Are Mot Met j MN RFO ! 6.6*83! fr.7us» ! *.S)04! 6.»611 «« ** * ! * !# , *b96 7.S *fl **** »» ! : 7.82*5 T 7.»752! MEAN * MODE» » , CFO ! X « » «»«»»*««**« #» «« t * * «< j MN *» «o » #» # ^ «» i MAX MAX STO ERROR COEF OF VARATON a 4 KURTOSS (8ETA 2) PEARSONAN SKE«* 0.12

89 DSTRBUTON LST No. of Copies Commander US Army Armament Command 12 ATTN: DRSAR-SAS 1 DRSAR-PP DRSAR-SF 1 Rock sland, L Commander Picatinny Arsenal ATTN: SARPA-MT-C Dover, NJ Commander Rock sland Arsenal ATTN: SARR-LPL Rock sland, L Defense Documentation Center Cameron Station Alexandria, VA Next page is blank. 87

90 >

91 1 - - ' *, > \» > si >'" v ' ' '. Ay /' ' % "- -t -, ^ '. "".. /' J \ V,'\ '1 ' es'- '/. /" A' /;' i ' v K, J, ' ' V t, '.v - r - V \. 1 ' 'V ' ' V - - N, r. i. r \ ', i ^' A 7.' ' >' 1* \ ; v - * > 'K- / X -. ) i / / jr/t i r y r,v V,. A ''"i'-.f" M A '. v \ V; ^ ^Pl ' [ ; ')»i ''!,- <-.' ^ *.'*AV r* ' ' * /' ')( V V A - /- ). t.1 "r 'V ' i t - tf ' ^ J - ' v V " 1 / \ '\ y j V ; -' ; / (».. * ' : ',. VA'

92 /. ',».^ $& r&rlv ' ~Y: S.'*;>*> >-.:- i V y. 7 / r ' 'v.' iv irfv^ V V'? \V k. ; *., >. "; y\_ ; i'ji V f t-*: \ i < / i r. / /';' V "" '"V s.v \.'* r - ; % vt. T ',... v 'Mti- * S V -'C- V,\*y 4"'^ " *-' ' )/ ; w 0. v> :>/'/ V. ' -'4>.? ', v ;>'';>- /^ ;'., r K.... 'k-v V -' / \ i. > >. ^ i ;-'^ V1 H' ~._V V, ' 7 ^v :-i; x >V - '-.4, S./ 1 kx % /., h 'v-.-v */ t '^'" " '" ' v; v i-u^ ; ^ «fi