Standard Specification or Quenched and Tempered Low-Alloy Structural Steel Plate with 70 ksi [485 MPa] Minimum Yield Strength to 4 in. [100 mm] Thick AASHTO DESIGNATION: M 313/M 313M-90 (ASTM DESIGNATION: A 852/A 852M-88a) 1. SCOPE JA This specification covers quenched and tempered high-strength Jow-alloy structural steel plates for ‘welded, riveted, or bolted construction, It is mended primarily for use in welded bridges and buildings where savings in ‘weight, added durability, and good notch toughness are important. The atmo- spheric corrosion resistance of this steel {is approximately two times that of carbon structural with copper (Note). Welding technique is of fundamental importance, and it is presupposed that the welding procesiure will be suitable for the steel and the intended service. This specifica tion is limited to material up to 4 in (100 mm}, inclusive, in thickness. [NOTE—Two times carbon structural steel with copper is equivalent to four times carbon svetral stel without copper (Cu 0.02 max). 1.2. Plates produced under this speci- fication shall be impact tested at a tem- erature not higher than 40°F [4°C]. 13. The values stated in inch-pound units or ST units are to be regarded sepa- rately as standard. Within th text, the SI ‘nits are shown in brackets. The values stated in each item are not exact equiva- lenis. Therefore, each system must be ‘sed independently of the other. Combin- ing values from the two systems may result in nonconformance with this speci fication, REFERENCED DOCUMENTS 24. AASHTO Standards: M 1601 M 160M General Require- ments for Rolled Steel Plates, Shapes, ‘Sheet Piling and Bars for Structural Use 22. ASTM Standards: AGI Specification for Gen- A6M_ eral Requirements for Rolled Steel Plates, Shapes, Steel Piling’ and Bar for Structural Use 4.370. Methods and Defini tions for Mechanical Testing of Steel Prod- ucts A 673/ Specification for Sam- A673M pling Procedure for Impact Testing of Structural Steel 3. GENERAL REQUIREMENTS FOR DELIVERY 3.1 Material fumished under this specification shall conform to the appli- cable requirements of the current edition of Specification M_160/M 160M. 4. PROCESS 4.1 The steel shall be made by one of the following processes: open hearth, basic oxygen, or electric fumace. 4.2. The steel shall be made to a fine grain practice S. CHEMICAL ‘REQUIREMENTS SL The heat analysis shall conform to the requirement preseribed in Table 1 52 When product analysis is. re TABLE 1 Chemica! Composition Requirements ‘Composition, Element percent Carbon 0.19 max Manganese 030-135 Phosphorus 0.04 max Sulfor (0.05 max Silicon o20-06s Nike! 050 max Chromium 040-070 ‘Copper 020-040 Vanadium 0.02-0.10 ‘quired, the steel shall conform on product analysis to the requirements prescribed in Table 1, subject to the product analysis tolerances in Specification M_ 160M 160M. 58.3 When required, the manufacturer shall supply evidence of atmospheric corrosion resistance satisfactory to the purchaser, 6. HEAT TREATMENT 61 The material shall be heat treated by the manufacturer by heating to tem- perature that produces an austenitic structure, but not less than 1650°F {900°C}, holding a sufficient time to at- tain uniform heat thrcughout the mate- rial, quenching in a suitable medium, and tempering at not less than 1100°F [593°C]. Heat testing temperatures shall be reported on the test certificates. 7. ‘TENSILE TEST REQUIREMENTS: 7A Tension Tests—The material as represented by the test specimens shall «9650 ‘SPECIFICATIONS FOR MATERIALS TABLE 2 Tensile Requirements! M313 Yield srength, min, ks OMPalP 70 (485) ‘Tensile swengt, range, ksi [MPa] ‘90-110 (620-760) Elongstion in 2 in. (50 mmf, min percent? 0 7 For plas wider as 26 i (610 de wat spaces ken ate wane doin, See Oneiaon under te Tension Tet eesti of MISO ISM «Foe thicoeses f 34 In (19 ron) ae uot, mets oa ei, (40-sm] wie fll icknes ectanglar specimen a shown in Figure of ASTM A S70, te anpaton readin 32 (50 mm) gage lenght inlodes ee fare and swe the rset lon For lar wier than 2 in 610ml, he songtionrequement i rediced to percentage pin "Foe boil hcenses under 5/6 in (8 mm) a delucton (om the speed peeeoge of elongation in (200mm shall e made for decreases of the nominal tickers blow 5/16 fri). Sov elon recuemea! ajo under te Tension Tes sect of M1684 60M fo sucion values. "Far nonna iclaezes ovr 35s. [0 ml Section fom the spss prema of elagation ia 2 in (90 am] shal be de for ners te nomi hicks shove 35k [90 ml Se slngaonrquenent sjuiments under the Tension Tens secon of MQM I6OM fr dedicion valor conform to the requirements specified in Table 2. 7.2. Number of Tests—One tension test shall be taken from a comer of each plate as heat treated. Plates wider than 24 in. (610 mm} shall be tested in the ‘wansverse direction and are subject tothe ‘modifications for elongation contained in footnote C of Table 2. 8. IMPACT TEST REQUIREMENTS ‘8:1 Longitudinal Charpy V-notch impact tests shall be made in accordance with test frequency H of ASTM A 673/ A.673M, By agreement, Charpy V-notch ‘impact tests may be made in accordance with test frequency (P) of A 67H/A. 673M. 82. The tests results shall meet an average minimum value of 20 fulbf (27 J] at 40°F (4°C]. By agreement, a test temperature lower than 40°F [4°C] or an energy level greater than 20 fLIbf, or both, may be specified, 9. RETEST 9.1 The manufacturer may rebeat ‘reat plates that fail to meet the mechani cal property requirements ofthis specifi- cation, All mechanical property tests shall be repeated when material is re- heat treated. ‘SUPPLEMENTARY REQUIREMENTS Standardized supplementary require- ments for use at the option of the pur- chaser are listed in Specification A 6/4, 6M. Those that are considered suitable for use with this specification are listed below by title, Sl. VACUUM TREATMENT, 82, PRODUCT ANALYSIS, $3, SIMULATED POST:WELD HEAT TREATMENT OF MECHANICAL TEST ‘COUPONS, 86. DROP WEIGHT TEST, $8. ULTRASONIC EXAMINATION, AND SM4, BEND TEST.i 1. SCOPE 1.1. This specification covers exter tullytieaded ste anchor ols in tended for anchoring structural supports to concrete foundations. Such stectural supports include building columns col umn supports for highway signs, street lighting, and trafic signals; steel bearing plates; and other similar applications. 142. The anchor bolts are subject to teasile property requirements, and aze available in three grades (yield strength levels). Special chemical restrictions apply when weldable steel is required. 13 Zinc coating (galvanizing) 1e- quirements are included fr bolts reqir- ing comosion protection. 4 The recommended grade and sole of nuts foreach grade of anchor bolt are included. 1.5. This specification does not cover the requirements of mechanical expan sion anchors or powder activated nails or stds for concrete. 1.6 The values stated in Suits are to be regarded as the standard 2. REFERENCED DOCUMENTS 2A AASHTO Documents M 160° ‘M 160M_ Specification for General Require- ments for Rolled Steel Plates, Shapes, Sheet Piling, and Bars for Structural Use M 232 Specification for Zine Coating (Hot Dip) on Iron and Steel Hard- M 291 Specification for Car- bon and Alloy Stee! Nuts M 293 Specification for Hardened Stee] ‘Washers M 298 Specification for Zine Standard Specification for Steel Anchor Bolts AASHTO DESIGNATION: M 314-90 Mechanically Depos- ited on Iron and Steel T 244 Methods and Defini- tions for Mechanical Testing of Steel Prod- 22. ASTM Standards: A751 Methods for Chemical Analysis of Steel Products 23° American National Standards: ANSI Unified Screw Bl. Threads ANSI. Square and Hex Nuts BI8.22 3, TERMINOLOGY SPECIFIC TO THIS STANDARD 341 anchor bolt—a steel bar of rod, fone end of which is intended to be cast jn concrete, while the opposite end is threaded and projects from the concrete, for anchoring other material to the con- crete. 3A dliscussion—The end of the bar or rod cast in concrete may be either straight or provided with an anchor such as a hook or welded or bolted attachment toresist forces imposed on the bolt, a re- quired. 3.2 fabricator—the manufacturer of the boli 33 producer—the manufacturer of the steel rods or bars. 34 purchaser—the purchaser of the finished anchor bolt, or his designated agent. 3.5 supplier—the agent who fur- nishes the finished bolts and nuts to the purchaser; may be the fabricator. 4. CLASSIFICATION 4.1 Anchor bolts covered by this specification are classified as follows: 4.1 Grade—The anchor bolts shall bbe furnished in one of three grades (Grade 36, Grade 55, or Grade 105) corresponding to the minimum yield strength of the steel (See Section 8.1). 4.1.2 Weldable stcs] shall meet the ‘requirements of Supplementary Require ‘ment S1, when specifically ordered. 5. ORDERING INFORMATION 54. Orders for anchor bolts under this specification shall include the fol- lowing information as necessary to ade- quately describe the desired product: 5.1.1 Quantity (number of pieces), [NOTE 1—if the purcheser intends to per- ‘orm destructive tess on fabricated bolts, the fabricator should be advised so that an ade- quate number are prodvsed, especially for sizes and grades not really availble from sock. S12 Name of material (steel an- chor bolts), 5.13 AASHTO designation includ- ing year of issue, SuL4 Grade of steel (Section 4.1.1), 5.15 Weldable stecl to meet Supple- ‘mentary Requirement $1, when required, 5.1.6 Dimensions—diameter (based ‘on nominal thread diameter), bolt length, thread length, length of hook (if hook is required), 54.7 Zine coating or other protec- tive coating (when required) (Section 64), SATA _ Specify the zine-coating pro- cess, whether hotdip, mechanically de- posited, or no preference, 5.72 Specify length from exposed ‘end to be costed, or eatire length, 5.1.8 Number of auts, either total ‘number or number per bolt, 5.19 Number of washers, either to- tal number or number per bolt, and di- ‘mensions if other than standard washers, S.110 Specify if inspection at place 651652 SPECIFICATIONS FOR MATERIALS M34 ‘of manufacture is required (Section ‘Tensile Properties 2», Elongation’, Miimam, ‘SAL Specify if certified test report erent is required (Section 14.1), and os ee 5.112 Any special requirements Mio, Moa Tensile Seng, 18200 mm S0mm_—-Minimam, Grade Ki) Min, MPa (ksi) (in) Gin) Percent 6. MATERIALS AND 3% «MB GH «HO WSIGEWS) 20. B 20 MANUFACTURE 55-3965) SIT 6SSTS 9) 18 21 30, 10s 724 (405) 862 w HORE (25 to 150) I 15 4 6A Bolts: "Ts reqvenen for elongtion i 200m (a) spline obs and bas eed alse. The earn G11 Steel for anchor bolts shal be ‘made by the open-hearth, basic-oxygen, cr electric furnace process. 6.1.2 Threads of anchor bolts may be rolled or cut, and shall conform to the requirements in Section 9.2. 6.1.3 Hooks on anchor bolts, when required, may be made by cold bending ‘or hot bending, provided no cracks de- ‘velop in the bend length. When hot bend- ing is elected, at no time shall any portion of the anchor bolt reach a temperature in excess of 595°C (1100°F). The bend- ing shall not significantly reduce the cross-sectional area of the bolt (see Sec- tion 9.3) 62. Nuts: 621 Nuts shall conform to AASHTO M 291 for inch-series threads for the grade and style indicated for each grade and size of anchor bolt, and ‘whether plain or zine coated, as shown in table entitled “Nuts.” 6.2.2. The requirements for the rec- ommended grade and style of nut may be fulfilled by furnishing @ nut of one fof the grades or styles specified in AASHTO M 291 that has equal or higher proof load stress. 62.3. Nats to be used on coated an- ‘chor bolts may be tapped oversize. The diametral amounts for oversize tapping indicated in AASHTO M 291 as mini- ‘mums are the maximum limits for over- size tapping for this anchor bolt specifi cation, NOTE 2—Steel to ste! thread engagement {s reduced between bolt and nut threads when the nut is tapped oversize. The purchaser should therefore be aware that muts zine ‘coated and tapped oversize will have alower tensile capacity than nuts that have not been tapped oversize. (See Table 3, AASHTO M 2), 63 Washers: 63.1 Washers, when required, shall for elongation in 50 mm (2 fm) split 0 est on machine specimens Bats ovr 80 3 men (to over 6 fo 102 tt (3 tof) 18 perce mn iy. 22 pret, over€3 10 76m (2; to 35a), 20 percent, “Anchor Bolts Size, mm (in) 36 6 w38 (hw ND over 38 10 102 (over 1; 1 4) 55 6 w38 (10 1) Nus pls Nuts Condition Gade ‘Sule Phin or Zine Coated ASH Plain or Zine Coated =A “Heavy Hox Phin A He 55 603801) Zine Coated A Heavy Hox 55 over 3810 102 (ver M104) Plain ot Zinc cowed © A Heavy Hex 105 61038 (0 1) Plain D Hex 10S 6 0 38.CH10 1%) Zine Coated DH Heavy Hex 105 over 38 t0 76 (over 1h 93) Plain or Zinc Couted__—DH_Heavy Hex conform to AASHTO M 293 unless oth- cerwise specified in the order. 64 Coatings for Enhanced Corro- sion Resistance GA Zine Coatings, Hot-Dip and Mechanically Deposited: GA.L1 When zinc coated anchor bolts, nuts, and washers are required, the purchaser shall specify the zine coating process, for example hot-dip, mechani cally deposited, or no preference. 6A12 When hot-dip is specified, the fasteners shall be zinc coated by the hhot-dip process in accordance with the requirements of Class C of AASHTO M 232. 64.13 When mechanically depos- ited is specified, the fasteners shall be zine coated by the mechanical deposition process in accordance with the require- ‘ments of Class 50 of AASHTO M 298. 64.1.4 When no preference is speci- fied, the supplier may furnish either a hhot-dip zine coating in accordance with AASHTO M 232, Class C or a mechani- cally deposited zinc coating in accor- dance with AASHTO M 298, Class 50. 64.2 Coatings other than hotdip and mechanically deposited zinc coat- ings, when required, shall be specified in the order and the complete specifica- ‘tion for the coating shall be included as part of the erder. 643. Threads for coated anchor bolts shall meet the requirements of Sec- tion 9.2 before the coating is applied. 7. CHEMICAL COMPOSITION Steel shall conform to the fol- lowing chemical limitations: Hest Product ‘Analysic_ Analyst Phosphorus, > percent cos 004s Sulphur, max pacent 0050 0.058 ‘Copper, min percent (en Ca is specified) 020s 72. The additional chemical limita- tions in Section $1.5 apply when bolts are ordered as weldable. 8.1 Steel bars from which anchor Dolls are mace, as well as the finished anchor bolts, shall conform to the tensileMaid requirements shown in table entitled “Tensile Properties.” 9. DIMENSIONS 94 Nominal Size—The nominal an- hor bolt diameter shall be the same as the nominal diameter of the threads 92 Thread —Dimensions—Threads ‘on anchor bolts shall be Unified National (Coarse Series as specified in ANSI BIL1, ‘nd shall have Class 2A tolerances. 9.2.1 Bolts to be used with nuts ‘which have been tapped oversize, in ac~ cordance with AASHTO M 291, shall have Class 2A threads before coating. ‘After coating, the maximum limit of pitch and major diameter may exceed the Class 2A limit by the following amount: Oversize Diameter, Limit, ‘am Gin) mm (nyt Dpto 1A Ch) inet 041 (0.016) Oier ILI 1 254 Cy ‘© Dy inl 053 (002) over 25.4.) 0:79 (0031) These values are the same a the minimum overapping required for zinc-oated nuts la AASHTO M 291, 922 The gaging limits for bolts shall be verified during manufacture ot use by assembly of nut tapped as nearly as practical tothe amount oversize shown above. In case of dispute 2 calibrated thread ring gage ofthat same size (Class X tolerance, gage tolerance plus) shall be used. Assembly of the gage or the Auk described above must be possible ‘ith hand effort following application of light machin oil to prevent galling and damage to the gage. Length of threads shal be the pied length, pls 28 mm in), minos zero 93° Bar Diameter: 931 When threads are rolled, the diamcter ofthe bar before threading shall 1otbe less than the minimum pitch diam- Gter specified for Class 2A threads in ANSI BILL 93.2 When the threads are cut, the diameter of the bar before threading shall ot be less than the specified diameter les the tolerance in AASHTO M 160/ M 160M for round bars 933. The bend section of bent bars shall have a cross-sectional area not less 90 percent of the area of straight SPECIFICATIONS FOR MATERIALS 653, =i Anchor Bol It with Hook Tn ‘aight Anchor Bolt bh = length of hook 1 = leagth of bolt {= leogth of threads (exposed end) y= length of threads (encased end), when requited 1 = length of zinc coating, min, when partial zinc coating is required FIGURE 1 Anchor portions of the bar. The area in the bend shall be calculated by the formula: As= 025 2D -d cross-sectional area in the bend. ‘minor (or minimum) diameter at any point, generally in the plane of the bend), ‘major diameter, atthe same cross- section as, and at 90 degree to, the ‘minor diameter, 9A Bolt Dimensions: 9A The overall length of straight anchor bolts, or length to the inside of the hook, shall be the specified length plus or minus 13 mm (¥/ in.) for bolt lengths of 600 mm (24 in.) or less, and plus or minus 25 mm (I in.) for longer bolts (see Figure 1), 9.4.2 The length of hooks shall be the specified length, plus or minus 10 percent of the specified hook length, ot Bott dimensions plus or minus 13 mm ('; in.), whichever is greater (see Figure 1). 9.43 The bend angle of hooks shall be the specified angle plus or minus 5 de- grees. 98 Coated Length: 95.1 When only the exposed end of the anchor bolt is required to be zinc coated, the length of zinc coating shall bee at least equal to the length specified in the order. There is no limit on extra length of zine coating, 10. NUMBER OF TESTS AND RETESTS 10.1 The steel producer shall. con- duct tests for tensile properies and chem- ical analysis and shall cerify the results to the fabricator of the anchor bolts, The fabricator shall be responsible for tests for zine coating and dimensional require- ments. 10.2 The purchaser srall have the654 right to perform any of the inspections and tests set forth in this specification ‘Tests may be made on either the bar stock or on the completed anchor bolts. 10.3 One test for chemical composi- tion shall be made from each heat. One test for tensile properties shall be made for each diameter of each heat, and if the bars are heat treated, for each furnace lot. The number of tests for zinc coating shall be as specified in AASHTO M 232. 10.3.1. Where identity to a specific hheat number (and furnace lot number for heat treated bars) has not been main- tained, the number of tests shall be based ‘on the quantity of bolts of a given de- scription, as shown in the table entitled “Number of Tests.” Number of Tests Number of Specimens 254 mm Number of Bolts (1 in) Over 25.4 mm in Order of Eich dia. and (1 in) Descition __smuller__ ia 100 and less 1 1 101 to 250, 1 2 251 to 500 2 3 501 to 1000 3 5 1001 and over {additional per 1000 bots) 1 2 10.4 If a sample fails to meet any tensile property requirement and the re- sults are within 14 MPa (2 ksi) of the required tensile stength, within 7 MPa (1 ksi) of the required yield strength, fr within two percentage points of the required elongation, a retest shall be per- mitted on two randomly selected speci- ‘mens for each original tensile specimen failure from the lot If ll esults of these retest specimens meet the specified re- quirements, the lot shall be accepted. 10.41 If any test specimen shows defective machining, it may be discarded ‘and another specimen substituted, 105 If a sample fails to meet any requirement, other than a tensile property requirement, and the results are within 10 percent of the limit, a retest will bbe permitted on two randomly selected specimens for each original non-con- forming specimen from the lot. If the results of tests on these retest specimens, for the requirement that was originally non-conforming, mect the specified re- uirement, the lot shall be accepted, ‘SPECIFICATIONS FOR MATERIALS 106 Ii a sample is determined 10 have threads not conforming to the re- quirements, where the determination was ‘made using other than standard thread ‘gages, and the fabricator contests the rejection of the anchor bolts or nuts, the final determination of acceptability shall bbe as follows. The fabricator shall have fullsize tension tests made on the anchor bolt and nut assembly in the presence of the purchaser's representative. The assembly must develop a tensile oad at least equal to the stress area of the bolt multiplied by the minimum tensile strength for the grade of anchor bolt specified. The suess area is calculated from the equation: 0.7854 [D ~ (0.9743in)} stress area, mm? (in?) nominal diameter of bolt, mm (in), and n= threads per mm (in), iL. TEST METHODS 11. Chemical analyses shall be per- formed in accordance with ASTM A 751 112 Tensile property tests shall be conducted according to AASHTO T 244 With the following exceptions: 11.241 Tension test specimens shall be the full section ofthe bar a rolled, except that bars over 38 mam (I's in) diameter in Grade 36 oF Grade 58, or over 32 mm (I in) diameter in Grade 105, may be machined to provide stan- dard 125 mm (0.500 in) specimens as described in Table 5 of AASHTO T 244 ‘when equipment of sufficient capacity for fullsize testing is not available. In the event that bolts aze tested by both fallsize and by machined specimen methods, the fullsize test shall gover if a controversy between the two meth- ods exists. 113 Determine the quantity of zinc coating by the methods of AASHTO M232. 12. INSPECTION 12.1 If the inspection described in Section 12.2 is required by the purchaser, it shall be specified in the inquiry and contract or order. M314 122. The inspector representing the purchaser shall have free entry to all parts of the fabricazor’s works that concern the ‘manufactore of the material ordered, The fabricator shall afford the inspector al reasonable féciities to satisfy him thet the material is being furnished in accor. dance with this specification. All tests and inspectioas required by the specfica- tion that are requested by the purchaser's representative shall be made, if possible, before shipment, and shall be conducted so as not to interfere unnecessarily with the operation of the works. 13, REJECTION AND REHEARING 13.1 Material that fails to conform to the requirements of this specification may be rejected. Rejection should be reported to the fabricator promptly and in writing. Incase of dissatisfaction with the results of the test, the fabricator may ‘make claim for a rehearing, 14, CERTIFICATION AND REPORTS OF TESTING 14.1 When specified by the pur- chaser, a certification that the materia ‘was manufactured and tested in accor- dance with this specification shall. be furnished an¢ shall include a report ofthe tensile test results and chemical analysis, Gncluding carbon equivalent for weld- able bolts). The report shall include the name of the steel producer, AASHTO. designation rumber, grade, heat number, size, and authorized signature, 15, PRODUCT MARKING 15, Bars rom which the anchor bolts are made shall be identified as specified in AASHTO M 160/M 160M. 15.2 The end of each fabricated an- chor bolt intended to project from the cconcrete shall be painted to identify the rade as follows: Grade Paint Color 36 Blue 55 Yellow 105 Redi i i i | | ] | M34 SUPPLEMENTARY REQUIREMENTS ‘The following supplementary require- ‘ment shall apply only when specified in the purchase order or contract: si, BOLTS SUITABLE FOR WELDING S11 The material described in this section is intended for welding. This sup- plemental section, by chemical composi- tion restrictions and by a carbon equiva- lent formula, provides assurance of wel- bility ‘$1.2 Welding technique is of funda- ‘mental importance when balts produced to this supplementary section are welded. Itis assumed that suitable welding proce- dures forthe steel being welded and the intended service will be selected. SL3- The requirements of this sup- plementary requirement supersedes con- flicting provisions of the general specifi- cation SPECIFICATIONS FOR MATERIALS S14 Because of the embritling ef- fects of welding temperatures on cold- forged steel, this supplemental section is limited to hot-forged bolts, of, if not forged, to the thread bars, studs, or bolts produced from hot-olled bars without forging. Cold-forged bolts or cold drawn threaded bars are suitable if they are given a thermal treatment by heating to a temperature of not less than 815°C (1500°F) and air-cooled. SUS Chemical Composition: S15.1 Steel shall conform to the fol- lowing limitations: Heat Product, Analysis Analysis Carbon, max percent 030 033 Manganese, max percent 135° Lal Phosphorus, max percent 0.040 0.048 Solfur, max percent 0.050 0.058 Silicon, max percent 050 055 655 SL52 Carbon Equivalent—In addi- tion to the requirements in Section $15.1, the analysis shall be such as to provide a carbon equivalent (CE) meet- ing the following requirements: $15.21 For alloy ot low alloy sttel, the carbon equivalent shall not exceed (0.45 percent when calculated as follows: Mn , %Cu , %Ni cE= 8c BG , BN a 40 0 BO Mo KV w- 0 0 $152.2 Forcarbon steel, the carbon ‘equivalent shall not exceed 0.40 percent when calculated as follows: Mn ce=ac+ 2h S16 Marking—Each fabricated an- chor bolt conforming to this Supplemen- tary Requirement $1 shall be designated by a white paint mark on the side of the bar near the end to be encased in con- rete.Standard Specification for Joints for Circular Concrete Sewer and Culvert Pipe, Using Rubber Gaskets AASHTO DESIGNATION M 315-94 AASHTO M 315-94 is identical to ASTM C 443-858, 656i Standard Specification for Joints for Circular Concrete Sewer and Culvert Pipe, Using Rubber Gaskets [Metric] AASHTO DESIGNATION M 315M-94 [AASHTO M 315M-94 is idemtical to ASTM C 443M-85a, 657Standard Metrie Practice Guide AASHTO DESIGNATION: R 1-91 Except for the following foreword, AASHTO R 1-91 is identical to ASTM E 380-89, but is not idertical to ASTM E 380-93, FOREWORD ‘To continue to serve the best interests of the public, the American Association of State Highway and Transportation Officials is actively cooperating with other standardization organizations in the development of simpler and nore universal metrology practices. The Standing committee on Engineering and Operations on November 14, 1975, reaffirmed the policy to gradually replace U.S. Customary units with those of the Intemational System of Units, generally known as SI. Recognizing the world- wide trend toward the use of this modemized metric system, the Association considers it important to prepare for its broader use by providing these guidelines for the use of SI in its standards through to coexistence of these two major systems. The Preferred and Standard dimensions with appropriate tolerances shall be stated first, Alternate measurement values, shown in parentheses, shall be rationalized rather than an exact mathematical conversion, whereas to rationalize is to round completely a converted value to a popular standard figure compatible with non-critical components, interchangeable parts, or other nominal sizes in a series. This policy involves no change in standard dimensions, tolerances, or performance specifications. To protect the new system from degradation, and to cooperate with knowledgeable people all over the world who are recommending good unit use, this standard strongly urges the use of SI, plus a very limited group of non-ST units, the need for ‘which is strong and is widely endorsed. Accordingly, while this standard agrees with the major standards in the world on the description of SI, it differs from some in the list of additional units. While all the units recommended in this standard are also recommended in the national documents of other countries and in ISO 1,000 (10}, some of the units recognized in these documents are not preferred for U.S. use, since suitable SI units are available for the quantities involved. 658. |1, SCOPE 1.1. This recommended practice is to provide guidance in the use and applica- tion of statistical procedures inthe preci sion analysis of sampling, testing, and scceptance of construction materials, 2, GENERAL PROVISIONS 24 It is recommended that AASHTO specifications for materials and for sampling and testing incorpo- rated accepted statistical standard proce- ures. The following ASTM designations provide guidance in this area and their latest published revisions are adapted as, a recommended practice forthe develop- ‘ment of AASHTO standards. Standard Recommended Practice for Statistical Procedures AASHTO DESIGNATION: R 4-82 21 ASTM C 670 Standard Rec- ‘ommended Practice for Preparing Preci- sion Statements for Test Methods for Construction Materials 2.2 ASTM C 802 Tentative Rec- ommended Practice for Conducting an Interlaboratory Test Program to Deter- mine the Precision of Test Methods for Construction Materials. 213 ASTM D 2915 Standard Method for Evaluating Allowable Prop- erties for Grades of Structural Lumber. 2.14. ASTMD 3665 Standard Prac- tice for Random Sampling of Paving Ma- terials, 2.8 ASTME 105 Standard Recom- mended Practice for Probability Sam- pling of Materials. 2.1.6 ASTME 122 Standard Recom- mended Practice for Choice of Sample Size to Estimate the Average Quality of Lot or Process. 24,7 _ASTME 141 Standard Recom- ‘mended Practice for Acceptance of Evi- dence Based on the Results of Probabil- ity Sampling. 21.8 ASTME 177 Standard Recom- mended Practice for Use of the Terms Precision and Accuracy as Applied to Measurement of a Propery of a Material, 2.9 ASTME 178 Siandard Recom- mended Practice for Dealing with Outly- ing Observations. 2.1.10 ASTM E 206 Standard Defi- nitions of Terms Relating to Fatigue Test- ing and the Statistical Analysis of Fa- tigue Data 24.11 ASTM E 456 Standard Defi nitions of Terms Relating to Statistical ‘Methods. 24.12 ASTM G 16 Standard Rec- ‘ommended Practice for Applying Statis- tics to Analysis of Corrosion Data, 659Standard Recommended Practice | for | Selection and Use of Emulsified Asphalts AASHTO DESIGNATION: R 5-89 AASHTO R 5-89 is identical t ASTM D 3628-93a except that all references to the ASTM standards contained in ASTM 1D 3628-93a, listed in the following table, shall be replaced with the corresponding AASHTO standard. Refer dard ‘AASHTO M0 D207 M 206 6601, SCOPE, 1. This recommended practice is to provide guidance in acceptable proce- Gures for measuring and monitoring the quality of air 2, GENERAL PROVISIONS 21_It is recommended that the fol- {owing Environmental Protection Agency Standard Recommended Practice for Monitoring Air Quality AASHTO DESIGNATION: R 7-81 procedures and their latest published pro- Visions be used in the monitoring of the uality of air. 2LL EPA-600/9-76-005, “Quality ‘Assurance Handbook for Air Pollution Measurement Systems, Vol. I, Princi- ples.” 2.1.2 EPA-600/4-77-027a, “Quality Assurance Handbook for Air Pollution Measurement System: Air Specific Methods. 2.1.3 EPA-600/4-77-227b, “Quality Assurance Handbook for Ait Pollution Measurement Systems, Vol. IT, Station ary Source Specific Methods.” Yol. 1, Ambient 661Standard Recommended Practice for ‘Transportation Related Earthborne Vibrations AASHTO DESIGNATION: R 8-81 Specification deleted due to lack of use. 662| | | | } Standard Recommended Practice for Acceptance Sampling Plans for Highway Construction METHOD A: PERCENT DEFECTIVE 1, SCOPE AND LIMITATIONS LA This recommended practice pro- vides guidance in the preparation of sta- tistical acceptance plans for transporta- tion materials and construction items. It is primarily intended to be used by AASHTO Technical Sections in the de- velopment of specification limits for AASHTO Standards where percent de- fective is the appropriate approach. 1.2. The acceptance procedures cov- cred by this guideline are based on PER- CENT DEFECTIVE (or its counterpart, PERCENT WITHIN LIMITS) as the ‘quality parameter. If it is desired to de- ‘elop an acceptance procedure based ona statistical parameter other than PER- CENT DEFECTIVE (such as the MEAN or STANDARD DEVIATION, for exam ple), then this guideline does not apply. (See Method B for alternate procedures based on the PERCENT WITHIN TOL- ERANCE,) 13 The various discussions, tables, and examples in this guideline assume that the acceptance procedures will be used to either accept or reject the work. While itis logical to extend this method logy to the development of acceptance procedures with adjusted pay schedules, that application is not addressed in this atideline, 2. GENERAL PROVISIONS 21 Requirements Plans: 21.1 The primary objective is to communicate in a clear and unambiguous ‘amner exactly what is desired. A partic~ lar specification requirement must be Xplicit and subject to one, and only ne, interpretation by the engineer, the of Acceptance AASHTO DESIGNATION: R 9.90 contractor, the materials supplier, the ac- countant, and the lawyer. 2.1.2 "To avoid potential legal dis- putes, the contractor should be given most ofthe responsibility for controlling the construction process while the high ‘way agency should be primarily respon- sible for judging the acceptability of the finished work through the use of properly designed acceptance plans. 2.13 The acceptance plan should be realistic in defining acceptable (AQL) and rejectable (RQL) quality levels. The AQL should be set high enough to satisfy design requirements but not so high that extraordinary methods or materials wll be required. The RQL should be set low enough s0 that the decision to reject is truly justified 2.1.4 All acceptance plans based on fractional sampling and testing are sub- ject tothe risk of making incorrect deci- sions. It is the responsibility of the de- signer of an acceptance plan to control the tisks at suitably low levels. 21.5. The choice of sample size must strike a balance between practical limita: tions on the number of samples that can bbe processed and the need for larger samples to obtain more precise informa tion (lower risks). 21.6 Mathematical and statistical principles must be applied correctly but the actual mechanics of the acceptance procedure should be simple and straight- forward, not requiring an understanding of the underlying theory 2.1.7 The plan must be suitable for use by the highway industry. It must bbe applicable to the various types of ‘materials and construction procedures ‘and must be adaptable to varying rates of production. 22° Measure of Quality Used in this Guideline: 2.2.1 The measure of quality used in this guideline is the lat percent defective. ‘This represents the percentage ofthe pop- ulation (lot) that falls outside a single lower or upper specification limit, or out- side both lower and upper limits, as illus- ‘tated in Figure 2.1. This term is well es tablished in quality assurance language and serves to tie Standard R9 tothe origi- nal source documents and standard text- books. Itshould be noted that an item hav- ingaasuitable level of PERCENT DEFEC- TIVE is considered to be entirely satisfactory (see definition of ACCEPT- ABLE QUALITY LEVELin Section4.4), 22.2 Altematively, quality may be fined in terms of PERCENT WITHIN LIMITS which isthe courterpart of PER- CENT DEFECTIVE, obtained by sub- twacting PERCENT DEFECTIVE from 100. An argument offere¢ in favor of the term PERCENT WITHIN LIMITS is that ‘thas a more positive psychological con- notation, 2.23. The use of either PERCENT DEFECTIVE or PERCENT WITHIN LIMITS is appropriate when these pa- rameters effectively describe the proper- ties that determine the ultimate perfor- ‘ance of the product. 224 This procedure does not pro- pose or recommend the incorporation of ‘materials, or acceptance of construction ‘components, that will impair serviceabil- ity or established safety standards, FIGURE 21 lustration of the Con- ‘cept of Percent Defective 663,664 225 The described procedures per- rit the selection of an ACCEPTABLE (QUALITY LEVEL which considers the impact ofa controlled maximum percent defective upon serviceability and safety 23 Sequence of Steps in the Devel- ‘opment of an Acceptance Plan 23.1 First, the specification writer smust decide thatthe panicular constuc- tion item i of sufficient importance that the effort and expense of a formal acoep- tance procedure is waranted. 23.2 Next, the characteristic (or characteristics) to be contolled must be chosen, Ideally, these should be strongly related to performance. Examples are strength and thickness of pavement and compressive strength of concrete 233 An appropriate test_ method rust be selected. This mast likely will be a standard procedure established by AASHTO, ASTM, or some other recog- nized agency. This is discussed further in Seation 43. 234 A specific statistical parameter (orparameters) must then be chosen upon Which the acceptance procedure will be based, This particular guideline adresses ‘only the PERCENT DEFECTIVE param- eter_(or its counterpart, PERCENT ‘WITHIN LIMITS). This parameteris de- fined in Article 22.1 235. In order 19 develop an accep- tance provedure using this guideline, the ACCEPTABLE QUALITY LEVEL must be defined in terms of PERCENT DE- FECTIVE. This step requires judgment ‘based on experience with the ‘materials and processes concemed. Sections 4.1 and 44 provide farther guidance 236 Although not absolutely neces- sary, it is often useful to also define @ REJECTABLE QUALITY LEVEL. This also requires judgment, is dependent upon experience, and is discussed in Sec- tions 4.1 and 44. 23:7 \tmustbe noted thatthe defini tions of ACCEPTABLE AND RE- JECTABLE QUALITY LEVELS in terms of PERCENT DEFECTIVE ae in- timately tied in with the SPECIFICA- TION LIMITS that are used. These are discussed in Anicles 4.1.2 and 4.1.3. 238 Once the appropriate quality levels have been defined, consideration must be given to suitable levels of RISK of incorrectly accepting unsatisfactory ‘work (BUYER’s RISK) or incorrectly te- ‘SPECIFICATIONS FOR MATERIALS jecting good work (SELLER’s RISK). This is discussed in Section 5.3-5.6, 239° Next, the specification writer rust decide between te two generaltypes of acceptance plans addressed by this guideline—ATTRIBUTES OR VARI- ABLES. The features and relative merits of these plans are discussed in Section 3. 23.10 Depending upon whether an ATTRIBUTES or a VARIABLES plan is desired, either Table A or B in the Appendix is consulted to determine the required sample size. This is explained in Section 7, and several examples are given in Section 12. 23.11 After the required sample size hhas been determined, the size of the lot from which the samples will be taken rust be defined. There usually is consid crable flexibility in making this decision as discussed in Section 6, 23.12 At this stage, the practicality of the entire acceptance procedure should be reexamined. Among factors to be con- sidered are the estimated benefit of qual- ity versus the manpower and other costs required to achieve it, the ability of the laboratory to handle the required number of tests, the ability of contractors to meet the specification, ete. If necessary, some of the original choices (LOT SIZE, AC- CEPTABLE and REJECTABLE QUAL- ITY LEVELS, BUYER’s and SELLER’s RISKS) can be revised and alternate plans compared, 23.13 If the methods in this guide- line are followed, and the basic theoreti- cal assumptions of distributional form and random sampling are reasonably well satisfied, then the acceptance plan can be ‘expected to perform as indicated by the OPERATING CHARACTERISTIC ta- bles in the appendix. However, before 2 plan is widely adopted, one or more field trials are usually advisable to eliminate any potential administrative or opera- tional difficulties 3, TYPES OF ACCEPTANCE PLANS 3.1 Acceptance Plans for Inspection bby Attributes: 3uL1 Acceptance plans for inspec tion by attributes are used when it is {impractical or undesirable to measure the specific value of a quality characteristic and, instead, each inspected item is clas- Ro sified as either satisfactory or unsatisfac- tory. The typical application involves the counting of some type of defect, such as the number of chipped bells in a lot of concrete pipe 3.1.2. Attibutes procedures may also be used with continuous data when the failure to satisfy the normality require- ment precludes the use of variables plans 33 Like the variables acceptance procedure desribed in Section 3.2, ate butes plans are designed to contol the percent defecive of the population (ot) However, thee are the following dis- tinct differences: 331 The decision reached is ae ceptor reject, no estimate of lot percent defective is made. 3.13.2 The population being sam- pled need not be normally distributed ay it must be for variables procedures. 3433 The attibutes procedure is Jess. Statsiclly efficient, requiring a larger sample size for a given diserimi- nating power. 3.34 An attributes plan may be based on a sige sample, if desiced. The variables plars described in this standard require a sample sizeof atleast n 32 Acceptance Plans for Inspection by Variables 32.1 Acceptance plans for inspec- tion by variables usually aply to quality characteristics that can be measured rather than counted and involve the com- putation of statistical parameters 3.2.2. Like the attributes acceptance procedure described in Section 31, var- ables plans are designed to contol the percent defecive of the population (lt). Unlike the atrbutes procedure, an est- mate of percent defective can be made Which provides the basis for an accep! or reject decision. 323 As a general rule, variables plans are more statistically efficient than atwibutes plans, providing greater dis: criminating power fora given samplesize. 3.24 Fundamental statistical param ters requteé for the application of the variables procedure presented inthis se°- tion are the mean and the sandard devia- ‘ion, computed as shown in Equations 3.1 and 3.2, G1)Ro 2) in which: sample mean = sample standard deviation individual sample values x Ss x x = sample size 32.5 Because the variability of the population is being estimated by the stan- {ard deviation computed from the sam= ple, aeceptance plans ofthis type are of- fen referred to as “Variability-unknown”” plans. 3226. In order toestimate the lor per- cent defective, iis first necessary to com- pte one or two quality index (Q) values, ‘sing either or both of Equations 3.3 and 34, a8 appropriate. L a G3) Qe Ga) in which, Q = quality index X= sample mean 5S = sample standard deviation 1, U = lower and upper specification limits outside of which the material or work is defined to be defective 32.7 The estimate of Jot percent de- {fective associated with any specific Q value and sample size is obtained from ‘he table in Appendix C. It will be ob- Served when using this table that a sample size of at least a = 3 is required for this ‘ype of variables acceptance procedure. 328 For cases in which there are both lower and upper specification limits, both Equations 3.3 and 3.4 are used and the individual estimates of percent defec- tive are added to obtain the lot percent defective. (Note that it is not correct to ‘dd the individual Q values and then de- ‘ermine the percent defective.) 32.9 Nommality Tests for Variables Data: 329. basic assumption of vari- sbles acceptance theory is that the popt- lon loo) that is being sampled is nor- Sally distributed, Many consiniction characterises have been found to closely SPECIFICATIONS FOR MATERIALS approximate this distribution. However, before deciding to use a variables accep” tance procedure, an attempt should be made to verify that the assumption of normality is reasonably satisfied. Once this determination has been made, itis assumed to apply to all future production and further normality tests would not rou- tinely be performed. 32.9.2 It may be sufficient to visu- ally check histograms of construction data to verify that the population being sampled is approximately normally dis- tuibuted, or to rely on various published studies. 3293 If a quantitative procedure ‘capable of producing a conclusion at a prescribed level of confidence is desired, the chi-square goodness-of-fit testis rec- ‘ommended. The details ofthis test may be found in many standard statistical texts, 3.29.4 If the assumption of normal- ity is not met, two remedies are possible. A greater number of acceptance tests can be performed so that averages of two or ‘more tests can be used in place of individ- ual tests. This strongly increases the ten= dency toward normality. Alternatively, an attributes procedure can be used which is uninfluenced by the distributional form of the population. 33. Other Types of Variables Plans: 3.3.1 Variables plans may also be of the “variability-known” type, in which the standard deviation of the product is assumed to be constant and known in advance. However, since itis usually one of the objectives of a statistical specifica tion to encourage and reward a greater degree of uniformity, this approach may bbe counterproductive. Since conditions ‘warranting the use of the “variability Known” procedure rarely exist in the highway field, this method has not been ‘developed in this standard 3.32 Variables plans may also be ‘based on the range (R) asthe measure of variability and many satisfactory plans of this type are in existence. The steps necessary to develop acceptance plans based on the range are outlined in Appen- dix D. However, it is expected that the standard deviation method will be the ‘method of choice for most future applica tions for the following reasons: 3321 The standard deviation ‘method is more statistically efficient, re- uiring smaller sample sizes to provide 665 ‘an equivalent degree of quality assurance. ‘This tanslates directly into cost savings. 3322 The standard deviation bas better mathematical properties than the range for various subsequent statistical analyses that might be performed, 33.23 Because the standard devi tion is computed from all the data values, rather than the two exteme values, it is Tess sensitive than the range to possible outliers. 33.24 The stantard deviation method is administratively simpler, avoiding the need for the subgrouping rules advocated with the range method. 33.25 Most scientific calculators now enable the standard deviation to be computed with a single keystroke, negat- ing the previous computational advantage of the range. 4. GUIDELINES FOR DEFINING AND MEASURING QUALITY 4.1 General Considerations: 44.1 In the highway construction field, itis seldom possible to define a single level of quality that distinetly dif- ferentiates between satisfactory and un- satisfactory work. However, it usually is possible o define a desized level of qual- ity that is clearly acceptable, and another lower level of quality that is considered to be so poor that it would be totaly unacceptable. This provides two conve- nent reference points from which an acceptance procedure can be developed, 441.2 For purposes of this standard, quality is defined in terms of percent defective, ie., te percentage of material or work falling outside stated specifica tion limits 4.1.3 When statistical limits are de- rived from historical dsta, care must be exercised to assure thal the data reflect ‘the normal variations of the measured property. In some cases, current speci cation limits have evolved from nonran- dom sampling techniques (pethaps re- sampling) or they may represent different test methods. For example, when applying Marshall voids criteria, limits ‘based on the bulk specific gravity of aggregates would not be appropriate for apparent specific gravity666, SPECIFICATIONS FOR MATERIALS Ro 42. Characteristics Used for Accep- 42.1 A significant characteristic is one that directly affects the performance fr appearance of a material or item of construction. As such, it is an indicator of quality, 42.2 A practical acceptance plan should define quality in terms of realistic levels of percent defective of signifi cant characteristics. 423° If a well established perfor- mance-versus-quality relationship does not exis, or ifthe characteristic of inter- ‘est is not readily measurable by a stan- dard acceptance test, it may be necessary to base the acceptance plan on the per- cent defective of some other measurable ‘property that is correlated with the char- acteristic of interest 43. Selection of Test Method: 431 IE possible, standard test method should be used. Both the proce- dure and any necessary equipment must bbe described or cited in the acceptance specification 43.2 The test method must be prac- tical and economical. This involves con- sideration of equipment costs, level of operator sill required, he length of time required to perform a test, and the possi- bility of obtaining timely results when remedial action is possible 43.3 The test method should be both accurate and precise to qualify as the basis for acceptance decisions. 433.1 Individual test results will al- ‘ways stay from the true population pa- rameter to some extent. Ifa large number of tests could be performed on identically the same product, the test results would not be identical but would be spread out in some form of statistical distribution. 433.2 The term accuracy refers 10 the degree to which the mean of this distribution tends to coincide with the ‘rue population parameter, 4333. The term precision refers to the degree of dispersion of this distribu- tion. The concept of good precision im- plies a narrow distribution of replicate test results, 4334 A test method may be accu- rate without being precise, and vice 4335 A method with less inherent precision may be preferable to a more precise method if a large number of test results can be obtained quickly and inex- pensively. The overall precision im- proves as the square root of the number of individual test results that are aver- aged together, 4.3.3.6 For some types of tests, itis impossible or impractical to ever know the true value of the population parame- ter. In these cases, the accuracy of an individual inspector or laboratory can be approximately judged by comparing the individval results with the grand mean obtained from many participants per- forming the test under carefully con- trolled conditions. This is the basis for some laboratory accreditation tests. 433.7 A test method that is not ac- curate is said to be biased. Steps should bbe taken to avoid the potential for bias that might result from nonstandard equipment or procedures or from un- ‘rained operators 4338 The percent defective esti- ‘mated from a sample by the method escribed in Section 3.2 is known to be ‘an accurate (mathematically unbiased) estimate of the percent defective of the population being sampled 44 Definition of Quality Levels: 441 The acceptable quality level (AQL) and rejectable quality level (ROL) are specific population (lot) cheracteris- tis. 44.2 The AQL is that level of per- cent defective at or below which the work is considered to be completely ac- ceptable. 443. ‘The ROL is that level of per- cent defective at or above which the ‘work is considered to be sufficiently un- acceptable that repair or replacement is warranted. 444 The basis for these definitions may be the subjective assessment of a panel of knowledgeable engineers, his- torical records of quality received and subsequent performance, or theoretical considerations. 445. The establishment of the AQL and ROL should be regarded as a separate activity to be completed prior to the development of the acceptance plan itself for any consideration of the risks associ- ated with the plan. 5. ACCEPTANCE CRITERIA, RISKS, AND CRITICALITY S.L_ General Considerations: 5.11 The primary reason for includ- ing numerical limits in an acceptance plan is to assure conformance with d sign requirements. The expected benefits of more stringent specification require- iments must be carefully weighed against the probable increase in costs S12 The exact quality for which the buyer is willing to pay should be specified in terms of realistic and en- forceable linits of significant character. ists S13 The acceptance criteria must be stated in terms of lower, upper, ot both lower and upper limits of significant characteristics outside of which no more than a specified percentage of the work (or # maximum number of test results in the case of atwibutes acceptance plans) shall be allowed to fal. 5.2 Acceptance Criteria: S2A For attributes plans, the lot size, sample size, and maximum allow able number of unsatisfactory test results is specified. 522 For variables plans, the lot size, sample size, and maximum allow- able estimated percent defective (IM) is specified. (Because it is necessary 10 allow for random variability ofthe sam- pling process, neither the AQL. nor the ROLis likely 10 be suitable as the speci- fied maximum allowable estimated per- cent defective of the acceptance require- ment. A satisfactory value will usually be found to lie somewhere between the AQL and the ROL. This is illustrated in the examples in Section 12.) 5.2.3. Altematively, in place of maxi- ‘mam allowable estimated percent defec- tive, a variables plan witha single lower or upper limit may specify the minimum acceptable value (K) ofthe quality index (© computes from the sample in accor- dance with Equation 3.3 or 3.4, For dou bile-limit plans, i is mathematically pref erable and administratively simpler to use only the maximum allowable esti- mated percent defective (9). 53 Types of Risks: 53 Buyer’ Risk (B) is the risk of erroneously accepting unsatisfactory (ROL) work. 53.2 Sellers Risk (a) is the risk of erroneously rejecting satisfactory (AQL) work. SA Factors Affecting Criticali: SA. From the engineering view- point, each acceptance plan should be ‘based on the criticality of the measured property as it affects safety, performance,Ro SPECIFICATIONS FOR MATERIALS. cor durability. In addition, legal and con- ‘ractual requirements should be consid- qed. The word “criticality” has been elected for use in this guideline to ex- press the overall concept of relative im- porance of various factors. The factors fo be considered in determining criti- cality ae’ Sofety—danger to human life. ‘Serviceability—inconvenience and other consequences, including mili- tary, oF disruption of service or use of the road or bridge. Cost—for construction, control, and ‘maintenance or replacement. Legal and Contractual Reguire- ‘ments—clarity of directions and in- terpretations of requirements, re- sponsibilities and obligations; rec- cognition of customary practice; fair and equitable bidding and accep- tance procedures. $5. Criticality Ratings: 55.1 For classification purposes, the following ratings of criticality are sug- gested: CCritical—when the requirement is es- sential to preservation of life ‘Major—when the requirement is nec- essary for the prevention of substan- tial economic loss. ‘Minor—when the requirement does ‘not materially affect performance. Contractual—when the requirement is established only to provide uniform standards for bidding. 5.6 Guidelines for Choosing Risks According 10 Criticality: 5.6.1 A large buyer's risk indicates ‘high probability of acceptance for poor (ROL) or marginal quality work. Al- ‘hough this may encourage lower bid Prices, it greatly increases the likelihood for prematare failures and higher mainte- nance cost. 5.6.2 A large sellers risk will result in the unnecessary rejection of satisfac- ‘ory (AGL) work and may eventually drive both the quality and the price 10 unnecessarily high levels, 5.6.3 It is the responsibility of the highway agency to design acceptance plans that control the risks at suitable levels appropriate for the degree of erti- cality of each application. The table enti- ‘Ned “Probability Values” gives probabil- Probability Probably of| Acceptance at RQL Classification (Buyers Risk) Gites 0.08 Major 0.050 Minor 100 Contractual 0200 ity values that have been suggested for specific degrees of criticality, although it is recognized that i may not always bbe necessary or practical 10 match these idealized values. ‘5.64 Analyzing the risks associated ‘with any particular acceptance scheme is usually accomplished by constructing ‘operating characteristic (OC) curves. Methods to develop OC curves are dis- cussed in Section 7. 5.6.5. Either or both the buyer's and seller's risks may be specified in select- ing a suitable acceptance plan although, because the sample size is a discrete variable, it may not always be possible to obtain precisely the desired risk levels. 6. SELECTION OF LOT SIZE AND SAMPLE SIZE 6.1 ‘The selection of lot size is dic- tated primarily by practicality and conve- rience although, for variables acceptance procedures i is important to avoid com- Dining two distinctly different popula- tions. The variables procedures described in this standard assume that the lot being sampled is a homogeneous normal distri- Dution. Attributes acceptance procedures ‘make no distributional assumptions and require less care in defining lot sizes. 62 Some practical limitations that Ihave been used to define maximum lot sizes are a day's production of concrete, a particular concrete item or structure, & Kilometer of roadway, a specified area or a day's worth of paving, a production run of a prefabricated or manufactured item, etc. 63 In the development of a statisti- cal acceptance procedure, the sample size is generally a more important consider- ation than the lot size because this strongly influences the risks involved and, consequently, the overall effective- ness of the quality assurance program. 64 The minimum sample size that 667 ves Probability of Seller's Risk Acceptance at AQL at AQL (0950 ‘0.050 0.90 010 0.995 00s 0999 0.001 may be used depends upon the type of acceptance procedure. The variables ac- ceptance procedures described in this standard require a sampie size of at least n = 3. Attributes procedures can be based on a single sample, if desired. 65 A more important consideration in the selection of sample size is the analysis of the risks associated with the acceptance plan. Only by constructing the operating characteristic curves, as described in Section 7, can it be deter- ‘mined if the sample size is sufficient to enable the procedure to properly discrim- inate between acceptable and unaccept- able work. 7. CONSTRUCTION OF OPERATING CHARACTERISTIC CURVES 7A Definition and Uses: Tal An operating characteristic (OC) curve is a graphical representation of an acceptance plan's ability to discrim- inate between acceptable and unaccept- able work. A typical example is shown in Figure 7.1 74.2 The importance of con structing OC curves cannot be overem- phasized. In this manter, the risks 10 both the highway agency and the contrac- tor can be determined in advance and, if necessary, modifications of the accep- tance plan can be made before trouble- some situations arise inthe field. Several examples are presented in Section 12. 74.3. The construction of OC curves for attributes acceptance plans is rela- tively straightforward and may be ac- ‘complished with the tables in Appendix ‘Acer the computational procedure in Ap- pendix E. JAA The theory underlying the con- struction of OC curves for variables ac- cceptance plans is considerably more complex and is beyord the technical668, SPECIFICATIONS FOR MATERIALS renga, TN oe sa “COO | 1 “I FIGURE 7.1 Typical Operating (Characteristic Curve for a Variables ‘Acceptance Plan scope of this standard. Consequently, only a tabular method is provided in Appendix B. An outline of the proce- dures used to construct the tables is given in Appendix E. 72 Operating Characteristic Ta- bles: 721 The tables in Appendix A and Appendix B provide the data necessary to construct OC curves for atibutes and variables plans, respectively 72.2 The tables span a wide range of potentially useful acceptance plans. Plans having relatively undesirable OC curves have been omitted and not all plans in these tables will be suitable for all situations. 7.23 Because attributes plans deal with discrete data, there is a finite mum- ber of practical plans, many of which are listed in Appendix A. The tables are constructed 50 that it will never be necessary to interpolate between accep- tance numbers (c) ot between sample sizes up to a sample size of n = 10. ‘Some interpolation may be necessary for larger sample sizes or for specific lot sizes, although the OC curves are rela- tively insensitive to lot size 7.24 Because variables plans deal ‘with continuous data, there are an infinite ‘number of possible plans that might be used. For either evaluating an existing plan or choosing a new plan, it may be appropriate to interpolate between the variables plans listed in Appendix B. 13. Procedure for Using Tables in Appendix A and Appendix B: 731 Select either Appendix A for attributes plans or Appendix B for vari- ables plans. 732 Select the risk considered more caitical (buyer’s or selle’s) and scan the appropriate table to determine those plans that approximately match this risk. 733 Determine from the tble which of these plans comes the closest to matching the less critical risk. 734 If necessary, interpolate be- tween plans to match the desired risks as closely as possible. 8. ATTRIBUTES PROCEDURES FOR SMALL LOTS Appendix A provides operating char- acteristics information for lot sizes down toN = 20. If necessary, acceptance pro- cedures for smaller lot sizes may be de- veloped in one of the following ways: 8.1 Specify an acceptance plan and use the hypergeometric formula in Ap- pendix E to develop the OC curve. Re- peat this procedure until a satisfactory OC curve is obtained. 82. Specify 100-percent inspection Ro and state the number of defective items that will be tolerated in the lot. 83° Specify 100-percent inspection and require that any defective items be repaired or replaced. 9. RANDOM SAMPLING 9.1 Importance of Random Sam. ling: 9.1.1 Of the various theoretical con- ditions upon which statistical acceptance procedures are based, the assumption of random sampling is one of the most {mportant. Only when all vestiges of per sonal bias are removed ean the laws of statistical probability be relied upon to fanction properly 9.2. Types of Random Sampling 9.2.1 Random sampling is often de- fined as 2 manner of sampling which allows every member ofthe population ot) to have an equal opportunity of appearing in the sample. 9.22 The mostbasie form of random sampling, sometimes called simple ran- dom sampling, allows every possible subset of the required sample size to have an equal chance of being selected ‘This is the lest restrictive definition but iit has the drawback that the sample loca tions occasionally tend to be clustered For this reason, most highway agencies use some method of stratification to as- sure that the sample locations will be spread more uniformly throughout the work 923 Swatified random sampling satisfies the fundamental requirement of Section 92.) while avoiding the clustes~ ing problem by dividing the lt into as many equal-sized sublots (stata) as there are samples to be drawn, A single ran dom sample is then obtained from each sublot 9.3. Random Sampling Procedures: 9.3.1 Suatified sampling plans for highway corstruction items are designed to spread the samples throughout the ‘work and tend to be quite similar. First, ‘most plans divide the lot into equal-sized sublots on the basis of ares, mass, oF ‘other appropriate measure. Then, within ‘each sublot, provisions are made to select ‘single random sample. A typical exam1 | is ‘SPECIFICATIONS FOR MATERIALS (NUMERALS 1-5 INDICATE SAMPLING LOCATIONS) 300m 300m 300m B00m SOOM. 24m 4L 0 5 a a 0 x 1500 m 669 DETERMINATION OF RANDOM X COORDINATES ADDITION TERM SAMPLE RANDOM = MUTIPLICATION TERM (CUMULATIVE LENGTH NUMBER NUMBER _(SUBLOT LENGTH) ‘To THIS SUBLOT) x 1 0.603 x 300 +0 181 2 0992 x 300 + 300 598 3 0.086 x 300 + 600 626 4 214 x 300 + 900 964 5 0551 x 300 + 1200 1365 DETERMINATION OF RANDOM ¥ COORDINATES SAMPLE RANDOM = MUTIPLICATION TERM NUMBER NUMBER (PAVEMENT WIDTI) XX 1 0750 x73 5S 2 0.286 x73 20 3 osa2 x73 40 4 81 x73 os 5 087 x73 65 ple of this approach is illustrated in Fig- lure 9.1. The uniform random numbers between zero and one are obtained from standard tables or may be generated by computer. 93.2 In practice, some agencies carry this method one step further. In sampling bituminous concrete, for exam- ple, it may be more convenient to sample rectly from the appropriate tucks rather than to wait until after the material thas been placed. In this case, the random locations in Figure 9.1 are used to deter- mine which trucks are to be sampled. ‘This is normally done in advance based on known total quantities and truck ca Pacities, 93.3 For the sampling of discrete items, such as sections of pipe, it may be desirable to devise variations of stratified random sampling. 10. RETESTING AND CORRECTIVE ACTION 10.1 Retesting Provisions: 10.1.1 When the original tests indi- cate that a large or costly item of work may warrant rejection, it is common practice to perform a series of retests before making the final decision. This constitutes a sequential acceptance scheme and will be dealt with in more detail in a forthcoming standard. 10.1.2 The use of a retesting provi- sion, and the exact manner in which the retest results are processed, both affect the risks associated with’ the accep- tance procedure. 10.13 If 2 retesting provision is t0 ‘be used, it should be described explicitly in the contract documents. 10.2 Corrective Action: ied Random Sampling Procedure Applied to Highway Pavement 10.2.1 When the acceptance proce- dare leads to the conclusion that a partic- ular Jot of material or item of work is to be rejected, the specification must spell out whatever corrective actions are to be taken. These actions will usually take the form of options at the discretion Of the highway agency. 10.2.2 The most obvious corrective action is to require the contractor to re- ‘move and replace the defective material or item at no additonal expense to the highway agency. This will not always be a practical course of action, however, and other options must be provided. 10.23 Many highway agencies have elected to include pey adjustment clauses in the specification so that, at their op- tion, defective workmay be accepted at 2 predetermined level of reduced payment. 10.24 Another method of dealing670 SPECIFICATIONS FOR MATERIALS Ro | with rejectable work is to require the 11.3 Because the result described in the bounding lot sizes of N = 20 and contractor to submit a written plan for Section 11.1 may be challenged by 2 N = 100. Because it is possible that corrective action to be accomplished at contractor who is unfamiliar with its the- there might occasionally be lot sizes no addtional expense to the highway retical basis, itis advisable to investi- smaller than N = 20, the OC curve for agency gate and reevaluate any lot rejected in alo sizeof N= 10 wil also be plot, 1025 As a last resort, it may be this manner. 12.13 The following information is necessary to negotiate a setement when obtained from Appendix A fora sample am item of work is determined to be re size of n = 10 and an acceptance number jectable. Aran of ¢ = 2 (see table “12.13") 121. Analysis of Arributes: Accep- M1, A POTENTIAL PROBLEM = ance Plan for Precast Concrete Cen- WITH VARIABLES PLANS ter Barrier: a er 12.1 An existing acceptance plan Lat ‘1.1 Although such occurrences are for precast concrete center barrier re- ee eee Tare, it is possible when using variables quires that n = 10 sections be selected Defective yc = 20 Lot Size = 100 acceptance procedures that alot may be at random and that various dimensional | gpa judged rejectable even though none of checks be performed. The plan further 1p ie ose the individual test results falls ouside stipulates that, if more than c= 2 of 15 oss oss the specification limits. Provided no fun- the n = 10 text sections fil to mest the 20 on 068 damental sumptions (normal popula- prescribed dimensional tolerances, the 25 050 052 tion, random sampling, etc.) have been entire Jot must be tested at the supplier's 30 031 037 violated, this is a theoretically correct expense and that all defective sections 3s 017 oe result. The proper inference is that, based shall be replaced at no extra cost to the “0 0.08 ots ‘on the mean and standard deviation esti- highway agency. The maximum lot sie SS 903 eas ‘mated from the sample, the population is specified to be a single production 2 ea a percent defective is unacceptably lage. run or N = 100 sections, whichever is 3S on na 112 This same result may also be smaller. Thee is no minimum lot size ceaused by one of more outliers, test re- but itis known from past experience that Salts tat deviate unusually fr from the its acy smaller than N= 20. eZ ton erent dete rnorm due to some assignable cause such 121.2 To analyze this acceptance ve is not 2 contimucas variable; 8 equipment malfunction or operator procedure itt desired t construct the PEL Gatos fine, sila ‘esror. ee operating characteristic (OC) curves for continous 2s has been done in Figure ™ T T ‘12.1. Given the fact that most lot sizes oso | | fall between N = 20 and N = 100, these Taal two curves provide practical bounds for oD scrmaures_Accerrance puan the probabilty of acceptance for aay level of submited quality. (Fo lot sizes on sence sizes «eto | of N= 10 orfewer, the lot is 100 percent inspected and the OC curve plots a5 & eo accerrance woutens «+2 || Peale) pmoanmire Ni j 1 { 12.1.5 Although an acceptable qual- orn ese { ity level (AQL) has not been defined in ae 1 | fee leal scrapes wae at hs often een oso} - 1 used in highway applications is 10 per- [.! cent defective In this example, i i sen 2.30 ae Aor sizes that the probability of acceptance for a ee | | Tota is rly 10-prcen defeeive les ae t T between P = 1.00 and P = 0.94, de- sue | | pending upon whether the lot size is 7 T N 20 or N = 100, respectively. This oo a L would probably be regarded as satisfac: © 1% 2 30 40 80 60 70 80 90 100 tory, stnongh this decition would be to the highway agency. The purpose of Lor rencenr cereerive the OC curves is to provide the informa: tion upon which this decision can lo FIGURE 12.1 Operating Characteristic Curves for Acceptance Plan for Precast cally be made. Concrete Center Barrier Described in Example 12.1 1216 Another factor that must beon ay SPECIFICATIONS FOR MATERIALS - 1 + 090 a5 I 4 T wanusces aceeorance run | | | are sure sees LI rants aos | | | oso Foo Miinwenss ts + saccerrance ’ in ; j | | | : | (ea ore T t rT 20 INA ! Lot pencent ocrecrive FIGURE 122 Operating Characteristic Curve for Acceptance Pan fr Pavement ‘Thickness Deseribed in Example 12.2 considered is the amount of unnecessary retesting that can be expected to occur when the process is truly under control. It can be seen from Figure 12.1 that, if most of the lot sizes tond to run close oN = 20, then lots that are uly 10- percent defective will almost always be accepted and the requirement for 100- percent inspection will almost never be triggered. If the lot sizes tend to run closer to N = 100, then 100 percent inspection will be required about 6.0 per- cent of the time when the lot is 10 per- cent defective 12.1.7 Like the AQL, no ROL (re- Jectable quality level) has been explicitly defined for this acceptance plan. If, for ‘example, the highway agency considered the dimensions of the center barrier to be ‘of moderate but not eritical importance, it is possible that the lot would not be regarded as rejectable unless 50 percent ‘or more of it was defective, I this were the case, the present plan would probably be considered quite satisfactory since, at SO-percent defective, there is at most 2 P = 0.05 probability of acceptance. On the other hand, if 30-percent defective were regarded as seriously deficient, then the present plan might not be satisfactory since the probability of acceptance at that level of quality lies between P= 031 and P = 037, 12.1.8 In order to analyze how the plan will perform with the occasional lot sizes that fall between N = 10 and N = 20, the OC curve for 2 lot size of N= 10 has been plotted in Figure 12.1. ‘This appears as a step function because it represents 100-percent sampling. The smaller Tots are not likely to pose a problem because those having 10-percent defective or less will always be accepted and those having more than 20-percent defective will always be rejected and subjected to 100-percent inspection. 122 Analysis. of Variables Accep- tance Plan for Pavement Thickness: 12.2.1 The SI equivalent of an ex- isting acceptance plan for pavement thickness requires that n = 5 cores be taken at random locations from each 1500 linear meters of paving, similar to the diagram in Figure 9.1. The following ‘acceptance requirement is specified: = Design Thickness +45 (12.1) in which, X = sample mean 'S = sample standard deviation k = 06 (the acceptance constant) 12.2.2 To evaluate the suitability of this acceptance plan, i is first necessary to construct the OC curve, To do this, it is observed that the acceptance require ‘ment in Equation 12.) can be rewritten in the following form. = Design Thickness s k 22) In terms of the quality index (0), is seen to be of the form described in Section 5.23: Qek 23) 12.23 The following information is ‘obtained from Appendix B by interpolat- ing for k = 0.6 between table values of k= 0.5972 and K = 0.632 for a sample size of n = 5. For variables acceptance procedures, probability of acceptance is 1 continuous variable and has been plot- ted as such in Figure 12.2 Lot Percent Probability of Defective Acceptance 10 0930 2» on 30 oa70 40 0265 30 013 o oes, 7 010 12.24 Like the previous example, neither the AQL nor the RQL has been specified. If itis assumed that the high- Way agency would be willing to tolerate up to 10 percent of the pavement having a thickness less than the design value, this implies that the AQL is 10-percent efective, and the comesponding proba bility of acceptance is P = 0.93. Conse- quently, approximately 7.0 percent of the lots of this (presumably. satisfactory) quality will be rejectec by this acceptance procedure. However, if rejected lots are subjected to further testing before the fi- nal decision is made, and a 7.0 percent retestng rat is not cousidered objection able, this may be regarded a a satisfac tory acceptance plan. 12.2.8 At the other extreme, if 50 percent or more of the pavement had a thickness less than the design value, this might be considered unacceptable and ‘would imply that the RQL is 50 percenton. defective. Iso, itcan be seen from Figure 122 that, if such a defective pavement were constructed, the highway agency's risk of accepting it would be only P 0.13, 12.2.6 As in the previous example, the AQL and ROL values of 10 and 50- percent defective, respectively, have been assumed only for discussion purposes. It is the responsibility of the highway agency to decide what levels are appro- phate for each particular application 122.7 The acceptance requirement, given in Equations 12.1 through 12.3, is presented in the form of a minimum ac- ‘ceptable value (&) of the quality index (Q). Itcould also be given as a maximum allowable estimated percent defective (WN. To determine what the appropriate value would be, it is necessary to interpo- late in Appendix B to determine that a minimum acceptable value of k = 0.6 ‘corresponds to a maximum allowable es- timated percent defective of M = 29.1 For practical purposes, this would proba- Dly be rounded off to M = 29. The accep- tance procedure utilizing this approach ‘would read as follows: 122.7.1 Take nm = 5 randomly lo- cated cores and determine the corres- ponding thickness values. Compute the ‘mean (X) and standard deviation (S) and, from these, the Q value in accordance with Equation 12.4 _ F Design Thickness 2 Ss 124) Consult Appendix C to determine the es- timated lot percent defective (PD). For the lotto be judged acceptable, it is r- quired that PD = M ice, PD = 28. 122.72 This plan is identical tothe plan requiring that Q = k. Le. = 05. It is important to understand that, al- though the AQL might be considered to be 10-percent defective, the accepiance procedure must allow the sample estimate fof percent defective to be as large as PD = M = 29, This allowance takes into account a sampling variability. When the Tot is truly 10-percent defective, there is a small probability of P = 0.07 thatthe estimated percent defective willbe equal toor greater than PD = 29. Consequently, when the acceptance requirement is set ‘SPECIFICATIONS FOR MATERIALS at M = 29, there is a probability of P (0.93 that the lot will be accepted, as illustrated by the OC curve in Figure 12.2. 123. Analysis of Variables Accep- tance Plan for Air Voids of Bituminous Concrete: 123.1 The SI equivalent of an ex: isting acceptance plan for bituminous ‘concrete requires that, for each 10000 square meters (or less) of paving, » = 10 ‘cores are to be taken at random locations. ‘The percentage of air voids for each core is to be determined by a standard test. ‘The mean (X) and standard deviation (S) Of the test results are to be calculated and, from these, the quality indexes (Q) associated with lower and upper limits of L= 2.0 and U = 7.0 are to be computed ‘with equations 12.5 and 12.6. 125) 12.6) The percent defective estimates associ- ated with these two Q values are obtained from the appropriate tables in Appendix. C and added together to obtain the total Jot percent defective. For the lot to be judged acceptable, it is equired thatthe total lot percent defective (PD) estimated in this manner be no larger than M = 25, 1232 It is stated in the footnote of the tables in Appendix B that the accep- tance probabilities are approximately or rect for double-timit acceptance proce- dures such as this. Fora sample size of n= 10 and a maximum allowable esi- rated percent defective of M = 25, the following values are interpolated from the table: Lot Percent Probability of Defective Acceptance 10 0.960 20 68s 30 0x0 0 ons 30 0.025 12.33. It can be seen by scanning these values that tis acceptance plan will accept most lots having a percent defec- tive of 10 or less and will usually reject Ro Jots having much more than 40-pereent defective. This plan would probably be regarded as satisfactory. 124 Development of Autribwes Ac- ceptance Plan for Depth of Cover Over Reinforcing Steet: 124.1 It is known that the spalling of portland cement concrete bridge decks is attributable primarily to corrosion of the top mat of reinforcing steel and that a clear depth of cover of 50 mm (2.0 in.) for more is one method of preventing or delaying this form of distress. Instru- ‘ments utilizing magnetic flux are avail- able that can accurately measure the depth of cover through the hardened con- crete. It is desired to develop an accep. tance plan to assure that at least 90 per- cent of the reinforcing steel has a depth of cover of $0 mm (2.0 in.) or more, 1242 Ifonly a single-sided specifi- cation is writen to assure adequate cover over the reinforcing steel, this will en- ‘courage contractors to set the steel deeper than desired, reducing the moment arm from the neural axis and threatening the structural capacity of the bridge deck. ‘Consequently, an upper limit on the depth ‘of steels also required. Since the location of the steel can be controlled quite accu- rately with the proper number and type of chairs and ties, itis considered reason- able to set the lower and upper limits faily close to each other. It is decided to develop a plan that would assure that at Teast 90 percent of the steel lies between depths of 50 and 75 mm (2.0 and 3.0 in), or conversely, that no more than. 10 percent falls outside these limits. There- fore, by defirition, the acceptable quality level (AQL) is 10-percent defective. 12.43 Itis important to note that this ‘would be an unreasonable and unfair re= quirement if data existed showing that contractors were incapable of achieving sufficient coatrol to keep 90 percent of the lot within these two limits, For exam- ple, if depth of cover were normally dis- tributed witha typical standard deviation of ¢ = 6 mm(0.25 in.), and the contractor were capable of controlling the mean depth exactly at the midrange value of 63 mm (25 in), then the limits would fall atplus and minus two standard deviations ‘and the lot percent defective would be about 5.0 percent. This degree of control fon the part of the contractor would beRo See ieee eo eng a enmally distributed, the degree of con- Sees 1244 Whether or not the highway agency chooses to define a specific level of percent defective as ejectable (RQL), jt must recognize that larger values of lot peroent defective are very undesirable for fo reasons. First, as the percentage of reinforcing steel having less than 50 mm (2.0 in) of cover increases, 50 does the likelihood of premature deck failure and the attendant costly repairs. Second, if the reinforcing steel is placed too deep in the deck slab, the deck may not be able to develop the load-carrying capacity for which it was designed. Consequently, when developing this acceptance proce- ure, an OC curve that drops off rapidly 1s percent defective increases will be de- sire. 1245 Normally, because the depth ‘measurements are continuous and vari- ables procedures make more efficient use ofthe data, it would be desirable to use variables acceptance procedure, How- ever, for purposes of this illustration, it is assumed that the highway agency has examined many sets of data and has found the distributions of depth of cover over reinforcing steel to be quite skewed. Since variables acceptance theory as- suumes thatthe population being sampled is normally distributed, while attributes ‘acceptance procedures are uninfluenced by distributional form, an attributes plan 's considered appropriate for this applica- tion, 124.6 When an attributes type of ac- ceptance procedure is applied to continu ‘us data, the lot size is considered to be infinite, Accordingly, the last table in Appendix A has been constructed for an infinite tot size. 1247 Since measurements of depth of cover are nondestructive and fairly easy to obtain, itis decided to ry asample size of n = 20, For an acceptance number of = 4, the following values are ob- ‘ined for an infinite lot size in Appen- dix a: SPECIFICATIONS FOR MATERIALS Lot Percent Probability Defect of Acceptance 5 1.00 10 Agu) 096 15 og 20 063 23 ost 30 026 35 012 20 0.08 45 0.02 30 001 ‘The AQL of 10-peroent defective will bbe accepted approximately 96 percent of| the time. For lots having greater amounts ‘of percent defective, the OC curve drops quite rapidly, as desired. Decks having 40-percent defective will be accepted ‘only about 5.0 percent of the time while those with 50-percent defective or more ‘will almost always be rejected, 12.5 Development of Auributes Ac- cceptance Plan for Pavement Thickness 12.5.1 A highway agency wishes to develop an acceptance procedure for pavement thickness that is as uncompli- cated as possible, involving no statistical calculations or special tables. The pave- ‘ment will be considered satisfactory if at least 90 percent of it has a thickness ‘greater than the design value. Therefore, the AQL may be considered to be 10- percent defective and it is desired that this level of quality have arelatively high probability of acceptance. At the other extreme, if 40 percent or more of the pavement is less than the design thick- ness, it has been decided that this will be regarded as rejectable (RQL) and a correspondingly low probability of ac- ceptance is desired. 12.5.2 For purposes of this example, suppose that a seller’s risk of a = 0.05 and a buyer's risk of 8 = 0.10 are desired. ‘The corresponding acceptance probabil ties are P = 0.95 at the AQL and P = 0.10 at the RQL. 125.3 The requirement for simplic- ity dictates an attributes plan. When atti- butes acceptance procedures are applied tocontinuous data (thickness inthis case), the lot size is considered to be infinite By scanning the rows and columns ofthe last table in Appendix A, itis observed that a plan with a sample size of n 15 and an acceptance number of ¢ = 3 produces very nearly the desired risk 3 levels. (Because the sample size and ac- ‘ceptance number are discrete values, itis not possible fo match the risks exactly.) The following values are obtained: Lot Percent Probability Detective of Accepeance 10(AQL) 094 20 065 30 030 RQ) 009 30 002 12.84. Itcan be seen from these val- tues that the basic objectives have been well satisfied. A good quality pavement having 10-percent defective or less will have a probability of acceptance of at least P= 094. IF the pavement is 40- percent defective or more, the probability of acceptance will be P = 0.09 or less. 1235 The compleed acceptance procedure will require that n = 15 cores be taken at random locations within a specified lot size. Provided that no more than ¢ = 3 cores are less than the design thickness, the lot will be judged accept- able. 126 Development of Variables Ac- ceptance Procedure for Percent Passing 0.075-mm (No. 200) Sieve: 12.6.1 An acceptance procedure is to bbe prepared for @ crushed-stone base course. The percentage by weight of ma terial passing the 0.075-mm (No. 200) sieve is known to be a significant perfor- ‘mance characteristic. Experience has shown that bases having 7.0 percent or les of minus 0.075-mm {No. 200) mate- rial have performed well but bases ex- ceeding 10.0 percent of minus 0.075-mm (No. 200) material have poor stability and drainage and tend to be fost susceptible. For this example, it is «ssumed that an analysis of historical dat has shown the test results on minus 0,075-mm (No. 200) material to be approximately normally distributed with atypical standard devia- tion of about o = 1.0 percent 12.62 The informaton provided in this example is sufficient to develop a worksble acceptance plan but itis not in the most useful form. For the types of acceptance plans coveredin this standard, definitions of acceptable and unaccept- able quality must be stated in terms of the percentage of material falling outside some specification limit (or pair of lim-674 DSTeUTONG ger newts) BASED OWT FIGURE 123 Mlustration of Defini- tions of AQL and RQL for Acceptance lan for Minus 0.075-mm (No. 200) Ma- terial Described in Example 12.6 its). Instead, the information provided in Section 12.6.1 is presented in terms of two average levels of minus 0.075-mm (No. 200) material that experience has shown have produced satisfactory and unsatisfactory results, respectively. As a reasonable approximation, these average values can be associated with the typical standard deviation of = 1.0 percent by means of normal distribution theory to provide guidance in establishing both the acceptable quality level (AQL) and the rejectable quality level (RQL) in terms of percent defective, The acceptance plan will then perform as desired as long as the standard deviation is reasonably close to the typical value and, if conservatively designed, it should provide ample protec- tion even when the standard deviation is larger than usual 12.6.3 On the assumption that there is no need to impose a lower limit on ‘minus 0.075-mm (No. 200) material for 1 crushed stone base course, this will be a single-limit specification. A logical choice for this limit is 7.0 percent, the level of minus 0.075-mm (No. 200) mate- rial that is known to be clearly satisfac- tory. It is believed thatthe base will per- form well as long as 90 percent or more of the material has a minus 0.075-mm (No. 200) value of 7.0 percent or less. ‘Therefore, the AQL is defined as 10-per- cent defective above the limit of 7.0 per- cent. This isa relatively conservative def- inition because, even ifthe standard devi- ation were considerably larger than the ‘SPECIFICATIONS FOR MATERIALS typical value, there is litle chance that any of the material in the normal distribu tion representing AQL quality would reach the known critical value of 10.0 percent minus 0,075-mm (No. 200) mate- rial. The AQL is illustrated in the upper % to 000 000 G0 Oa GOD 0 wD a0 zd a0 so wo oe co 00 09 100 ¥00 oro zo oro 70 180 460 oot oor 8 oe co 00 09 0 10 00 Oro wo oro wo eo wot ot Ss ca oo 0 079 0 79 100 0 so Izo Oro wo wo ol > rs 0 000 09 0 9 md 100 fo #0 Ged so oo oo o @o 000 00 0 OO GD OO To zo “0D Br wo 0 |e ve to 10 wo so cro so oo uo so LO S60 wo ots st oo wo 1d 20 soo of0 oo HO Tso ocd sto %0 or st to 00 00 0 Wo S00 wo sO sd sro sO HO wo wl E st 0 00 od 0 oo 100 wo sO 110 Io so oO fo wo F st Wo wo $0 0 9D sco Leo 160 $90 BLO «FO S60 wo mI ¢ ov 0 000 wo wo soo Go sto sro sco wo wo eo wo oo Z ot fo 00 070 WO 10 wo wo wo sd ED 90 HO HO wo fT ot wo soo Goo sto so seo ao 190 m0 FO wo WO sD mI 6 0 00 wd wo wd wo zo to oro O70 wo uO wo mI F 6 0 %0 00 100 wo so 0 IO S10 Gro eo oO wo wo 1 6 50 ord 90 sco 9f0 wo oO Wo Ifo O60 seo B60 OI oT ¢ 8 od wo $00 800 ico to co sso 0 oFD O50 L6d OOT — z 8 oo 0 wo 10 0 S90 O10 sro HO oF OFo WO wo S60 T 8 20 0 600 0 wo ed wo cso svo sD SRO oO wo oT Zz “ 0 io wo #0 90 G0 sro wo GO pro iO wo oO 560 t so 10 urd sco s€0 wo wo S90 so PFO WO so so mI z ° 100 0 v0 90 O10 90 0 Io to FO 970 Bo oro wo I ° oro eo 10 OF oF0 WO HO WO HO O80 sso B60 HO mI z s @0 sd 0 ero #0 sco fo zo fo eo wo wo wo wo I 5 | 0 zo um Po Ico 60 uO 950 s¥o HO wo GRD sO oO I + | 10 too wo wo 900 90 wo iro fo eo ora wo so Wo 0 + | 0 ao so wo oFo S50 S90 LO GLO S80 C6 HO UO HO € | 0 wo S90 oo wo sro i2o io wo wo wo To wo oOo ‘ | wo cro 910 oo so oro 90 wo oro wo wo wo 80 oo 0 z | 0 sto oro sro oF0 sso O70 sv OLd SLO OF s#O O60 Soo 0 ' (ue eS ww SS SS we Ss OF Ss o oO saqumy as SSR ROSA 10H PARAS OY SEY PRO sundeny tus | ops as ior san soueNay SGVSPECIFICATIONS FOR MATERIALS 6 TABLE A. Operating Characteristics of Attributes Acceptance Plans—Continued utributes Acceptance Plans Lot Size ‘Genple Acseptance Probability of Acceplance for Selecied Levels of Lot Percent Defective ‘Nurnber sie pie)” 100 100 100 100 100 100 a dn Fe Pe Oo oe 100 7,00 0.09 0.00 0.00 0.00 009 009 0.00 0.00 0.00 Goo 00 00 1.00 1.00 0.00 0.09 0.00 0.00 0.00 0.00 0.00 0.00 0.00 000 000 00 100 1.00. 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 a0 a0 0.00 100 100 0.00 0.09 0.00 0.00 0.00 0.00 0.00 0.00 0.0) 00 0.00 0.00 100 1.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.0 000 00 0.00 1.00 100 0.00 0.00 9.00 0.00 0.00 0.00 0.00 0.00 0.00 00 0.00 0.00 100 1.00 0.00 0.00 0.00 0.00 0.00 0.00 0.90 0.09 0.00 a0 000 0.00 1,00 1.00 0.00 0.00 0.00 0.00 0.00 9.00 0.00 0.00 0.00 0.00 0.00 0.00 Auributes Acceptance Plans Lot Size 500 0.95 0.90 0.85 0.80 0.75 0.70 o 060 055 050 045 040 0.35 0.20 © 090 081 072 068 056 049 o 1 036 030 0.25 020 016 0.12 0.09 3 086 073 O61 051 042 034 022 017 012 0.09 005 0.08 0.03 3 099 097 084 0.90 O8 O78 0.72 065 057 050 043 035 0.28 0.22 0 O81 0.66 052 0.41 032 024 O18 0.13 0.09 0.05 0.04 003 0.01 0.01 1 099 095 089 082 0.78 065 056 047 039 O31 02 GI 013 0.08 092 084 0.74 063 053 043 0.36 0.26 0.19 013 009 005 003, 099 097 094 090 O84 0.77 068 059 050 O41 032 023 016 097 089 078 066 053 042 032 023 0.16 011 007 004 002 0.01 098 095 090 083 075 065 O56 044 0.34 0.25 O18 012 0.07 096 085 072 058 O44 035 025 O16 0.10 0.05 0.03 002 0.01 0.00 098 093 085 076 065 053 042 032 022 01S 009 005 0.03 094 081 066 050 0.37 025 0.17 010 0.06 0.03 002 a0 0.00 0.00 096 090 080 068 0.55 043 0.31 022 0.14 009 005 002 0.01 1.00 1.00 098 0.95 089 081 O71 059 048 036 025 017 0.10 0.06 093 0.78 060 0.43 030 0.19 012 007 0.0% 002 001 000 0.00 0.00 BS e 8 8 t 2 099 095 086 0.74 0.60 046 0.34 0.23 0.15 0.09 0.05 002 0.01 0.00 3 1.00 099 097 092 O84 073 061 048 036 0.25 0.16 0.10 005 0.02 0 1 092 0.74 054 037 024 O15 0.08 0.04 0.02 0.1 000 000 0,00 0.00 0 2 099 099 082 0.68 052 038 026 0.16 0.10 005 003 001 0.00 0.00 0 3 1.00 099 095 088 078 065 051 038 026 0.17 0.10 005 002 0.01 15 2 0.97 082 0.60 039 0.23 0.12 005 0.03 0.01 0.00 0.00 000 0,00 0.00 5 3 1.00 095 083 0.65 0.46 0.29 0.17 0.09 0.08 0.02 0.01 000 0.00 0.0 15 4 100 0.99 0.9% 084 069 051 035 021 0.12 0.05 0.02 001 0.00 0.00 is 5 1.00 1.00 0.98 094 085 072 056 040 026 0.15 007 003 0.01 0.00 2» 2 093 068 0.40 0.20 0.09 0.03 0.01 0.00 0.00 0.00 0.00 009 0.00 0.00 20 3 099 087 065 0.61 0.22 0.10 0.04 0.01 0.00 0.00 0.00 000 0.00 9.00 2 4 1.00 0.96 0.83 0.83 041 023 0.11 0.05 002 0.01 0.00 000 0.00 0.00 2 5 100 0.99 0.94 0.8) 062 041 024 0.12 008 0.02 0.01 000 0.00 0.00 2 6 100 1.60 098 092 079 O61 O41 025 0.13 0.05 0.02 001 0.00 0.00 0 3 995 065 031 0.12 0.03 001 000 0.00 0.00 0.00 0.00 000 0.00 0.00 ‘ 30 4 0.99 0.83 0.52 0.25 0.09 0.03 0.01 0.00 0.00 0.00 0.00 000 0.00 0.00 i 30 5 100093 O71 O42 0.19 0.07 0.02 0.00 000 0.00 0.00 000 0.00 0.00 ( 30 6 1.00098 085 O61 034 0.15 005 0.01 0.00 0.00 0.00 000 0.00 0.00 | 30 T 1.00099 0.94 0.77 O51 0.27 0.12 0.08 0.01 0.00 0.00 000 000 0.00 i 30 8100 100 0.98 0.88 0.68 0.43022 0.09 003 0.01 0.00 000 0.00 0.00 | __ebstiliy of accopance fsa fencton of for sine for abates plane wsing crete dat. For variable ft ses, wll Be aecosany 1 plot bounding Operating | racers carver The valu in the abe are appropri for Doh sugle-lait and double it apletons sod are arnloseed Oy We dations! frm af | be popaatn680, SPECIFICATIONS FOR MATERIALS Ro TABLE A. Operating Characteristics of Atributes Acceptance Ptans—Contnued Atributes Acceptance Plans Lot Sie = ‘Sample Acceptance Probability of Accepance for Seated Levels of Loi Percent Drfecive ‘Sie Number ——— — — a) © Si FLO 1s aise ast is) Oo ee a 30 5 097 062 021 008 01 0.00 0.00 000 G00 0.00 300 000 000 000 50 S 099 0.78 035 0.09 0.42 0.00 000 000 0.00 000 000 00 000 0.00 50 7 1.00 089 052 0.18 0.08 001 000 000 0.00 000 000 O00 000 0.90. 50 3 1.00095 067 030 0.08 00: 0.00 0.00 0.00 000 000 O00 000 000 50 3 1.00 0.98 080 0.88 0.18 003 0.00 000 0.00 000 000 OMn 000 0.00 50 10 L009 0.99 0.89 0.59 0.25 0.07 0.01 0.00 0.09 0.00 0.00 000 0.00 0.00 50 L009 100 0.95 0.72 037 0.13 003 0.00 0.09 0.00 0.00 000 0.00 0.00 100 0.96 6.30 0.02 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.99 009 0.00 0.00 100 0.99 0.44 0.04 0.00 0.00 0.00 0.00 000 0.00 000 4.00 000 0.00 0.00 100 1.00 0.59 0.08 0.00 0.00 0.00 0.00 0.00 0.00 000 0.00 0.00 0.00 0.00 100 100 072 0.14 001 0.00 0.00 9.90 000 0.00 000 000 000 000 0.00 100 100 083 022 002 0.00 0.00 0.90 000 000 90) 000 000 000 0.00 100 100 090 9.33 003 0.00 0.00 0.00 6.00 0.00 0.00 000 000 0.00 0.09 100 100 0.95 0.45 008 0.00 0.00 0.00 0.00 0.00 0.00 000 000 0.00 0.00 100 100 0.98 0.57 0:10 0.01 0.00 0.00 0.00 0.00 0.00 0.00 000 0.00 0.00 100 100 0.99 0.69 0.16 0.01 0.00 0.00 0.00 000 0.00 000 000 00 0.00 100 100 1.00 0.79 0.25 0.02 0.00 0.00 0.09 0.00 0.00 6.00 000 0,00 0.00 100 1.09 100 0.86 038 0.08 0.00 0.00 0.00 0.00 0.00 «00 000 000 0.00 [Adtributes Acceptance Pl Lot Size = Infinite 1 © 095 0.90 0.85 0.80 0.75 0.70 0465 060 O55 050 045 040 035 0.20 2 0 0.9 081 072 064 056 049 042 036 030 025 620 0.16 012 009 3 9 086 0.73 061 O51 0.42 036 027 022 0.17 013 COP 0.06 0.08 0.03 3 1 099 097 094 090 084 078 072 065 057 050 C4 035 028 02 4 © O81 0.66 052 041 032 024 O18 0.13 0.09 006 cos 003 002 0.1 4 1 099 095 089 082 074 065 035 OM 039 O31 C2 O18 013 008 5 1 098 092 0.84 0.74 063 053 043 034 026 0.19 «13 009 0.05 003 5 2 100 0.99 097 054 0.90 084 076 O68 059 050 G4 O52 O28 O16 6 1 097 0.89 078 066 053 042 032 0.23 016 0.11 007 004 062 0.1 6 2 100 098 095 080 083 074 O65 O54 Om 034 026 O18 O12 O07 7 1 096 0.85 0.72 058 044 033 023 0.16 0.10 0.06 008 002 001 0.00 7 2 1.00 097 0.93 O85 0.76 045 053 042 032 023 O15 010 006 003 8 1 094 081 066 0.50 037 026 0.17 O11 0.06 004 G02 001 0.00 0.00 8 2 0.99 0.9 089 0.80 0.68 055 043 032 022 O14 009 005 003 0.01 8 3 100 099 098 09% 089 081 O71 059 O48 036 02% 0.17 O11 0.06 ° 1 093 0.77 060 G44 0.30 020 012 0.07 00% 0.02 001 000 0.00 000 9 2 099 095 086 078 0.60 046 034 023 0.15 0.09 005 003 001 0.00 9 3 1.00 099 097 091 083 073 O81 O48 036 025 B17 010 005 0.03 10 1 091 0.7% 054 0.38 0.26 015 0.09 005 002 001 000 000 0.00 0.00 10 2 098 083 0.82 068 053 038 026 0.17 0.10 005 003 001 000 0.00 10 31.00 059 0.95 088 0.78 065 O51 038 027 0.17 010 005 003 O01 1s 2 0.96 082 0.60 040 0.24 013 0.06 003 0.01 0.00 000 0.00 0.00 0.00 5 5 099 0.9% 082 065 0.4 030 0.17 009 0.04 002 00 000 000 0.00 5 % 100 0.99 09s 088 0.69 052 035 022 0.12 005 003 001 000 0.00 5 5 100 1.00 098 094 08 07 055 049 026 015 008 003 O01 0.00 20 2 082 068 040 021 0.09 0.08 0.01 0.00 0.00 0.09 00 000 0.00 0.00 20 30.98 087 0.65 041 0.23 0.1 0.04 0.02 0.00 0.00000 0.00 0.00 0.00 robe of aectce a fncin of fo sre for abs pls wing Gace du. Foy varie Io snes, W wile pxemany pot Doueding pean farce curve, The vals th ue ste spropate fr bat Sing end abe lin ape sad‘SPECIFICATIONS FOR MATERIALS 681 Ro. ‘TABLE A. Operating Characterstis of Attributes Acceptance Plans—Contnued netputes Acceptance Plans Lot Size = Infinite Gonple Acceptance Probability of Acceptance for Selected Levels of Lat Perceni Defecuve a Number sie © @ $s 0 15 wo 3s sow a CT 024 012 005 002 001 0.00 0.00 000 000 aD 5 100 099 093 0.80 0.42 0.28 0.13 0.05 002 0.01 0.0 0.0 0.00 Be 6 100 1.00 098 0.91 061 042 025 0.13 006 002 0.01 00 0.00 094 065 032 0.12 x0 3 0.01 0,00 0.00 0.09 0.00 0.00 0.0 0.00 0.00 0 4 098 082 052 026 010 0.03 0.01 0.00 0.00 0.00 0,00 0.00 0.00 0.00 x0 5 100 0.93 071 0.43 020 0.08 0.02 0.01 0.00 0.00 9.00 00 0.00 0.00 x0 6 100 0.97 085 061 0.35 016 0.06 0.02 0.00 0,00 0.00 000 0.00 0.00 0 7 100 0.99 0.93 0.76 0.51 0.28 0.12 0.08 0.01 0.00 0.00 0.00 0.00 0.00 a 8 100 1.00 0.97 087 0.67 043 022 0.09 003 001 0.00 000 000 0.00 0 5 0.96 062 022 0.05 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 3 6 0990.77 036 0.10 0.02 0.00 0.00 0.00 0.00 0.00 000 0.00 0.00 000 0 7 100 088 052 0.19 0.05 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 50 8 100 0.94 067 0.31 0.99 0.02 000 0.00 000 0.00 0.00 0.00 00 000 0 9 100 0.98 079 048 0.16 0.0% 001 0.00 0.00 0.00 0.00 0.00 0.00 0.00 30 101.00 0.99 088 058 0.26 008 0.02 0.00 0.00 0.00 0.00 0.00 0.00 0.00 | 50 " 1.00 1.00 094 0.72 038 0.14 003 0.01 000 000 0.00 0.00 0.00 0.00 100 8 0.84 0.32 0.03 0.00 0.00 6.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100 9 0.97 O45 0.06 0.00 0.00 0.00 0.00 0.00 0.00 9.0 0.00 0.00 0.00 0.00 10 100.99 0.58 0.10 0.01 0.00 0.00 0.00 0.00 0.00 9.00 0.00 0.00 0.00 0.00 100 u 100 070 016 0.01 0.00 0.00 000 0.00 0.00 0.00 0.00 0.00 0.00 0.00 - 2 100 0.80 025 0.03 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 i 100 B 1.00 0.88 035 0.08 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 fi 14 1.00 0.93 0.46 0.08 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100 5 1.00 0.96 0.57 0.13 0.01 0.00 0.00 0.00 0.00 9.00 0.00 0.00 0.00 0.00 100 1 1.00098 0.67 0.19 0.02 0.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100 0 100 0.99 0.76 0.27 0.04 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 00 18 1.00 1,000.84 0.34 0.06 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Frbainy of acceptance a Fancion of lt sos forstibutes plans using cree Goa, For variable To ace, wll be pocesany plot Wundingopeaing ‘hance corer The values inthe tale are appoprise for bath singel an doublet aplication sad are wnnfesced oy the doar of De poplin682 ‘SPECIFICATIONS FOR MATERIALS. Ro | APPENDIX B OPERATING CHARACTERISTIC TABLES FOR VARIABLES PLANS ‘TABLE B, Operating Characteristics of Varables Acceptance Plans Yaviables Acceptance Pans__t**_Varabiiy Unknown Procedure__***_Standard Deviation Meta exinwe cow pecans oy Protaity of Acceptance for Sete Levelt of Lt PereentDefecve o ra wo «ot 3 3 ous 03; oons O15 or 3 a ous Oss 836 ons 3 Pa ors 098 st 0 bls ‘ 2 os 066 006 002 ‘ 2 ost om oor ne ‘ 2 oss one 00s ls ‘ Fr 098 oon oro oes ‘ 36 ost eit Oba ‘ 3 os” se ais 0s ‘ a aor oe cists : “ 09 oat 020 Oe ‘ % a3 O93 023 eto 5 2% os9 04s o10 = ott 5 2 ny os foal 3 x0 a3 ons ois Oost 3 2 03s on? ois oso 3 u 09 ois oer Om 3 36 oor oe ot O07 8 3 Ps om O87 023 or 3 2 09 990 02s ose 3 2 099 ose tar aus 3 & 0338s ox isa 6 x» ax oe cos 02a é 6 as O87 cas =o: 0 ‘ 2 03s on ose bis aos § Fa oss ne ot Omen é 2 097 ao ors aos é x O98 oe ote 906 at ‘ x oss Gar os (boom ‘ 2 033 osz om on om ‘ 2 10 ose ost bis eat 1 2 oss os7 oo oa 7 Pry 0.91 0.63 0.05 0.01 0.00 7 6 0.93 0.63 0.06 0.02 0.00 > 09s Ons cos aoe 00 > oor O78 aio 03a ; 09 Ose a2 age (Ot 7 0.98 0.36. ots 0.03 oot 7 36 0.99 0.89 0.18 0.06 oot ‘Te acceptance probablies inthis able we scoat [ov sgl plant and we approninmely comew for dooce oa. For singe mit plan, eer ‘maximurm allowable estimated pecem defective (mn) or He misimum allonale gealty inden th) ay be speed. For coubilit plan. oly the esi ‘Slowable exited percent defective soul be wedRo ‘SPECIFICATIONS FOR MATERIALS 683 TABLE B. Operating Characteristics of Variables Acceptance Plans—Continued rales Acceptance Plans__*"*_—_Variablity Unknown Procedure Standard Deviation Method Maximum Alowsble Minimum | se ee = (hm _ Probably of Acepune for Stet Levels of Lt Percent Defective ' os arr room Tanai | 7 Fy 099 ro on os 021 oor 0.02 | : Fs 10 089 Ons aaa ame | ‘ 2 090 058 028k, 0.03) | t 2 ox gs 03st ke 0 i ‘ 2s 09s 070s 8s | 3 2 09 «073 kae_ 820m i ‘ Fy 09 = 0880s mm | i 2 O58 seats } 8 34 0.99 0.87 0.62 0.33, O13 0.08 oor | : % 05 = 09ers ast | 3 38 Mo ee ye | 5 20 er ee | ° 2 yy Nee ty ty i 5 Fa 09 oes 22st i 3 x 0% «07 8s oe i ° Fs 0s 0% oes, i 3 40 058 = oe!) i 5 2 099 = o8s 088m | 5 3 0% = 08) eso | ° 36 roo 0928808 Os ae { 10 20 0.89 os oat 0.06 0.01 0.00 0.00 i ‘0 z os 059025 ams fb a 09 = «8D 10 25 097 = om 037s, io Fs Bop Wh te ten 10 xo 09 = 082 = 030 oa! ae io 2 09 = O87 0s? a6 008mm ” x 10 09s} OaT atta 5 10 1s x 2s h aie 00 © basse 2 U8 09 ome =a? 8 Tie os on aaa ‘om 09 ors oa? as 5 1037 097 = 078s asa 15 16 0396 ca elie 0s ee 07] Melos Rem rs " ose 099 08s oe? 15 is 03% 0% 088s) 5 » oses 0% 03a 13 2» oes 10 0s ost 030 3 Fa oats 19 085 ome 5 2 0 10 ose os? ad 15 2 oer 10 097 oss ome 8 X ons to 0st okra 6 Fa oucas to 0989 most sto is 2 os? 10 099 Ostome asset 3 n ozs to 09 Oss oro} 8 FA ose to 09 oss oa?sws tte ; » se ooh tet cole onc ati ss Bee Mtner PUMetnsc ease ; 20 10 ian 0.90 ose ons 0.09 0.03 0.01 0.00 i» 1219 093 0s aa) otek ; .» is ite 09s oe) 3s oles ‘Te ecpuce potatos is ule we accra Ter Sng nl plan sd te appovimey cone fr deal paw Fora i alow entrain dette (nthe mma wae ety et) ay be etd ar Swe pone Main i684, ‘SPECIFICATIONS FOR MATERIALS TABLED. Operating Characteristics of Varlables Acceptance Plans—Continusd Variables Acceptance Plans co Variability Unknown Procedure ‘Standard Deviation Method Maximum Allowable Minium Estimated Allowable one pean ee Probability of Acceptance for Selected Levels of Lot Percent Defective _9) © a 5 0 5 20 25 30 3s 2 3 ary 097 = os ane 2 16 Low 09% = 078s 022 ki oot 2 is 03s 098 = 08 sz 028 ttt 2 6 0.996 099088088 013 00st 2 0 0.936 09 08st 35 le 0.06 20 ie oat 10 092088 at ois = 008 2» 9 0.882 Me he ee 20 0 o.8as yoo 0957s st om 020.8 20 2 ogi ro (os? oe 055 31s as 2 2 om 000.98 oss oz 077 20 23 076 oo 098 fe, ee tconee suet om) eo! 20 2 0712 ro 039k om oa 0280.10 20 25 0.680 ro 0880s as sl oe 20 26 0.689 09 hoo 08s ays 0s 0 9 13m 09 04s 0030.00 0.00 30 10 Las ost = 033 le ae. 0.00 0 u Oe 0 en 0.0. 0 2 Lia 098 = 08 = 030m oaz 0.00 30 5 nia 099 078 = sT 203 0.00, 30 ia Lov 099 08 es 0.0 30 5 1036 = Op CN on 0.06 0.0 30 6 0.995 100 01 039026 O08 0.00 30 7 0.985 100 0930s ott 001 30 8 ooi7 10 098k oor 30 9 0.880 we Oo 8 0.01 30 2 oes 100 078 oss sk 2 0.02 30 2 0.809 10009807 os? 0.02 30 2 os 100099000633 0.03 30 2B 07 1.00 00 ae 003s to eet oa) 008 30 8 1398 090 0290s 00.00 0.00 50 9 1335 09s oat oo oa 0.00 0.00 50 0 iam os = 032m. 0 50 iu 1223 099 Os 017 og 0.00 0.00 30 2 1s 099 073 02% = 00s = 0.00 0.00 50 B uns oo os? 033005) 0.00 0 14 1.980 ro oss ooo 0.00 30 15 1036 too oss aa. 0.00 50 6 0.995 Koo 085 Osta 0.00 0 0 0.955 W320 (ey 0.00 30 18 0316 ro 08907 asa 0.00 50 9 0879 roo 0980s a. 0.0 0 20 oss 1004.00 odo 0.16 0.0 100 7 Lan 089 010 0.0000 0.00 0.00 100 8 402 oss 021 0 0.00 0.00 0.00, 100 ° Lae 1 er 0.00 100 0 1279 too esate 0321 euo.0 cee vonote | 0.00 0.00 100 " rns 100-068 08.00.00 0.00. 100 2 vim Loo 080 lsat 0.00 0.00 100 5 Ls 1000890251 0.00 0.00 100 4 080 ee 0 tte 9.00 100 1s 1036 100098 =o: 0.06 0.00 0.00 100 6 0.995 100099 oss ott 0.00 0.00 100 0 09st ttoofeeee 100) ee. olee es T0c101 eo 0.00 ‘he acepance probes is able a accurate franglo it pasa and we apponnsily cones fr dese plans ov sng imi i, ber smatimuetalowabeexinuted pertent defective () oF the munimor lovable guaty index) may be oeied. Fer Sosbe-tnt fa, ony Oe SS ‘Slowabie eames percom cfecve shld be wed,Ro ‘SPECIFICATIONS FOR MATERIALS 685 APPENDIX C TABLES FOR ESTIMATION OF LOT PERCENT DEFECTIVE TABLE C._ Estimation of Lot Percent Defective by Standard Deviation Method ‘Variability Unknown Procedure Standard Deviation Method cam ninaed La Paces Deena for Sled Sampe St ‘oO 3 4 5s 6 7 8 9 © » » 5» a0 30 530.00 3.00 Som 0 soo wm Jom m0 wo mo Ho sm on 82 96 Be OS 8 BO Be mH Be Be BO Be Heo om 84 33 #2 wr 2 | we 2 a Ba wal gal oD oo 17 G0 ts aso feat as as wees Me oaeet ake) al ak do 9 or st ss) 5) a7 thee ae Gees ea aa) abl te 8a a 2 wie 2 el wo woe wos ale ee ‘on dos $855 800 ates aa as ts ames er ate a1 to ao ae 4 oa aa as fas a a as ole om dm 97 4 Gas Hos Ol doar os Sess Mowe ae wae toa 0% «92 So M90 4% Soar Ges” Sosy ose Now MoaT sos wea O10 728 4567 54s 4639 4626 224618 6616 46.10. HOR eos 4603 on 465 4635 SG Soe Sas ses soe aN ass ee sa ok #68 6m Gm So so Se be So os 9 3m Sa Os feat sgt sae Sas Sle sos 0s So oe eo ata tae de 4 Gs Sm we an Gn fe ae 5S dest leet as dis 85 sme eas wo ass aaa] as ele eh ac fas dle a558 0 er aasr aes akon 62 Sh Gn on fn Se Be of den Gee G7 Bes 53 20 fy bs Sn f= Bh Oi Sm tw Be ao B28 65 SN bo a0 22 oe on 0 uM er as or Bor ean an ge 2s fs fa Ra 00 44s 02 a9 20 os a0 3 2% 219 ais a 2% on ie Bm Bs a ar fa ton ais fe tie i do om bm ae mis 9s aso Se Ne de ite fur ae far oa me as fee 4139 Stee iy ha ie fe fom es fo ox bs 2m us aie dior fas ‘ose “on anes ae 56 aoe das as er fis 40865090 o2 fo fs om on bis ins 0a a fis om fos0 4035 1s om 9s be Ba DD oT o” 2 fw as oe S997 bn em ws 350 sds a 3938 dm 2 ane 0s 8% 90 S40 a ly aoa ear ps 380 03 in a om bean a8 3903 or Gal 3875 hes Skea 030 4163 409 3937 9.063887 366 3860 3eae 3837 3831 38.26 3824 os 13s oer on samo 850 Seo bass Seo snop 93 nay 786 O32 ios os her eae ete bo Mae me sel mss sane 0335 fom pm asa ans 3778 355 ay Hondo Sle tas 0 Os fog er 9 See Sr Sis saa sean 368736803875 3602 03; fom ay Me snatn05 we Saas Ma) dod Seas sea 30s Oa Ses 3800 37.28 3551 3059 3955 MeN BS MD eI? esol SST Om aner 77 3695 5655 3653 od oon Seon Sas 875 Set eh 3800 Om 393 3733 Sesh 3620 Soe sss ses Ste sas SENN ae 388 03 39m ro 3625 S58 SR sa? ssT 3829) SSI SOL od MaDe 040 3674 36.67 35.88 5493526 35.1 3800 3493 HTS KS SE SD BaD ou eas 3635 Ssst Sa Moo as Sees es? Seared heae Se O42 akis 3600 sy be19 ass Mao eas Saat 3800 nae esa ane 0 374s ser has Mad hin Mor 3393 GaSe Se Gk oa 315 3553 Sas Mos eke Set SST Se) 9a Shs O45 i126 3500 Hie Sa SS Ba BD ee ME HT HTT Re Ou 68 eT Se) 38 HB hae Me Moe Ses a Ss aN oar 366s 343s har Shoe dame aaa Sete te Soars ‘utes in dy of able ae enimtes of fx percent defective comesponaing o ape valice of uaiy dex and sonple size For Q valves ener wo eal ‘he percent eleive estimate may be teas dedtly om the ble Far Q valus less than sco the ble ales mat be oot og 0686, SPECIFICATIONS FOR MATERIALS Ro, ‘TABLE C. Estimation of Lot Percent Defective hy Standard Deviation Method —Conthnued Variability —Unknown Procedure Standard Deviation Method, Quality Estimated Lot Percent Defective for Selcied Sample Sizes Tdex Rovio sos 6 2 8 9 1 ts 2 300 ttH os «635 MOO BR Ne Ba BI Bis Bor Nas ST Me 3.6 Shoo 049 3605 33.67 2783234 OB 5192 3180 51.72 3180 Hat 3333128 050 3575 3333 32.44 32.00 S74 SST BAS 31.37 SEIS 34.06 30.98 30.93 30.80 051 3544 33.00 IO 3165-3139 31.22 3110 51.02 39.00 0.7 3063-3087 30.54 052 S13 32.67 31.75 3131 S408 30.87 30.76 0.67 3045 3034 30.28 © 30.23. 30:9 05] M82 3233-3142 30.96 30.70 -30.53 30.41 30.52 30.10 30.01 29.93 29.88 29.84 05 S431 32.00 31.08 30.62 30.36 30.18 30.07 29.90 39.75 29.67 29.59 2953 29.49 055 34.20 31.67 30.74 30.28 3001 29.84 29.72 29.68 VAl_-29.52_— 9.28. 9S 056 35.88 31.33 40 29.94 © 29.67 29.50 29.38 20.29 79.07 28.98 28.90 7885 RAL 037 35.57 31.00 0.05 29.60 29.33 29.16 29.04 28.95.7873 6G RSE 2651 8.7 088 35.25 3067 29.73 2926 28.99 2882 28.70 28.61 28.39 28.30 28.22 817-2813 059 2.93 3033-2939 2895 28.66 ©2848 28.356 28.28 HOS-27.96—«27.89—-S—27.79 0.60 32.61 30.00 29.05 2359 28.52 28.15 25.03 912-2. VSS- 7027.86 061 3228 2967 mM «IS 98 BITS -|D—7.22_— IB MAS 062 31.96 2933 #397792 76S 4B 77.36 77.27 «ZS 696 26.89 2683 26.80 063 8.63 29.00 28.05 2759 -77.52 «71S 29.03 26.98 ©2672 26.63 28.56. 25.50 26.87 064 «31.30 -28.67 272-7726: -26.99-26EZ_ 26.70 26.61 26392831 26.2} PHIE 26K 065 3097 28.13.2739 2692 25.66 25.49 26.37 26.28 ©2607 2598 25.90 25.05 25.89 0.66 30.63 28.00 77.05 26.60 25.33 26.16 26.04 25.96 2574 2566 25.59 2553 25.49 067 3030-2787 26.73 26.27 25.00 2583 25.72 2563 2542 5.53 25.26 ISI 25:17 0.68 29.96 27.33 26.40 25.94 25.68 25.51 25.39 28.31 25,10 250 24.98 2489 24.86 069 29.61 27.00 26.07 25.61 25.35 25.19 25.07 24.99 24.78 4G) 2462 24ST. 070 29.27 26.67 25.74 25.29 25.03 24.86 2475 24.67 1.6L ME HSL 7426 4.3 O71 2892 2433-41 2496 © HTD 24S 24K 24.33 HIS 24.05. 28.09 73.98 28.91 0.72 2857 26.00 75.09 2464 4.392425 4:11 2403-2383 25.75 23.68 23.64 3.60 073 BR «567 2476 2432 2HOT-H91 3.80 28D HQ ME 7337-7333 3.30 07 «27186 15:33 2444 400 BIS 7359 3.49 BAL HDI_—2BIZ_«-28OT._«-73.02 22.99 075 1730 2500 6.11 368 BAe 2528 3.17 B.D 9D DH BIE NR 2 07% 73 Me BI B37 BR RT Dae DD Neo 2 BA WAZ 239 OM «267 433 B47 230 BE NE 2256 RM I 22 IG WIZ 209 0.78 2639 2400 BIS 2274 SO 3S IS BIS 2199 92 «186 LED 21.79 079 2602 B67 BE 242 DIY WOK 2194 287 2726S 2LS3_—21.50 O80 25.64 2333 BSL IL BS eM 21ST ADE LDF. 21T ost 25.25 23.00 BI9 2180-258 2146 «34 2027-21020 ©2098 20.94 20.92 082 24.86 22.67 2187-2149 21.27 214 21.04 20.98 ©2081 20.75 20.69 70.66 20.63 083 2447 2233-2156 2118 20.97 2080.75. 20.68 ©7052-2046 ©2040-2037 20:38 086 26.07 22.00 21.28 2088 20.67 2056 20.45 20.39 2023.20.17 20.12 20.09 20.08 08S 25.67 21.67 2093 2037 20.37 2024 2016-2010 19.9 19.88 19.64 19.80 19-78, 086 25.26 21.33 20.62 2027-007 1995 19.87 «9A 19.64 «19.60 19.56. 19.5319.51 087 284 21.00 20311997 19.78 1964 19.58 19:52 19.38 19.52 19.28 19.28 19.29 O88 © 242 20.67 20.0 19.67 19.48 1937 19.29 19.23 19.10 19.08 19.00 1898 18.96. 089 21.99 2033 1969 1937 19:19 19.08 19.00 18.95-«1882.«18.77 «18.73 18.70— 18.69 099 21.55 20.00 19.38 19.07 18.90 18.79 18.72 18.67 18.54 8.50 8G R18 O91 211 19.67 19.07 1878 S.6] IRSA 84k 1R39 ZT 1823818 092 2066 19.33 1871849 18.33.1823... 17.96.1792, 17.90 17.89, 093 2019 19.00 18.46 1819 18.08 1795 17-8078 17.73, 17.69 17.65.76 ITS 094 1973 186718161790 17.7676? 17.6756 KG T7738 17ST 095 1925 1833 1786761748 173917381728 17.29 AGATA TIT 0.96 1875 180017551733 :17.20«1712_—«17.08. «1703-68 16.916.8E_ 16 BT_—16.86. 097 1825 17.67 17251704 «1692-1637 16.79 16.75 16.68 16.65 16.63 16.6 16.6 098 1774 17.33 16.95 1675 16.65 1657 16.53 16.49 16-42 «16.39 16.37 16.36 16.36 099 17211700 16.65 16.48 16.37_—16.31_—16.26 16.23 16.16 16.18 161216 IAG. ‘Namber In boy af le we etic af prcetdfcave comsiponaing o specie vals of quay len and ple san Far Q valor ewer enor ial {0 too, the percent Geese exirate maybe read dey tom the ele For vac ese than the te vals ont be wstod fom 100Ro SPECIFICATIONS FOR MATERIALS 687 TABLE C. Estimation of Lot Percent Defective by Standard Deviation Method—Continued ‘Variability Unknown Procedure Standard Deviation Method Quality Estimated Lot Pescent Defective for Selesed Sample Sizes ‘nde @ 3 4s 6 1 SO 0 Go fo 100 1667 ©1667 16.36 16.20 16.10 to.0e 1600 13.97 58] sa) S88 IST ISR Fol 16.11 16331607 15.92 18.83 15.78 15.74 15.72 S66 1S.64 563 15.63 15.62 Ho. 1553 16.00 15.78 15.64 15.56 ISL 15481546 ISL 15.40 1539 15.39 1538 ros 14931567 S48 15.37 18.30 15.25 523 15.21 SIT ISIS ISIS ISIS ISIS Lo¢ 1431 15.33, 15.19 1509 15.03 15.00 16.97 14961492 491189) 1491 14.91 10s 13.65 1500 14.91 1482 14.77 14.74 1472 14.7116 1GT «KT 168148 106 1298 1467 1462 14.55 14ST 4491447141445 107 12.27 14.33 14.33 1429 14.26 1428 4.2) 14.22 4.20 a2] 2? 14.22 108 15114001405 1421800 13.99 1398 13.97 139713971398 13.99 148.00 109 10711367 13.76 1376 ISTE BTS B74 B95 B77 3.77 JO 9.86 13.33 13.48 13.50 13.49 13.49 1349 13.50 13S 3.521389 13.SA 13.55 889 1300 1320 13.24 1325 13.28 13.26 1336 1.28 132913311332 13:36 iu 12 782 12.67 12.93 1298 13.00 13.01 13.02 13.03 1308 13.07 1309 1AY TBI 13 660 1233 «1265127227577 2-79 2adRAB 28S AT 2A Lid 5008 12.00 12.37 124725). 125412581257 126 12.63 12.66 12.68 «1270 Kis 287 Ler 21022230232 19.34 391A SATA 116 0.001133 1183-1197 12.03 12.07 210 1212_—T «12.21 12.26.—1238 LIT 0.00 11.00 1561172 11.79 LBs 1487 19.90 11.96 12.00«12.03«1206 1208 118 0.00 10.67 291671156 LNG) 6S 1.68175 11.7914.82 11.85.88 1190.00, 10.532 1.231333 tN Se TSB 8116S 168 1.20 0.00 1000 1075 1099 10 ITB 121 0.00 9.67 10501075. 108710981099 OSS IB 1216 ED 12 0.00 9.33 1023 1051 10.65 1073.10.78 1.21093. 1098 L.OT 1.09 123 0.00900 9.97 1028 104210511057 10.61 10.73 10.78 10.84 1088-1091 126 000 867 9.72 10.0 10.20 1030 10.36 10.41 10.53.1059 10.64 10.69 10.72 125 0.00 8.3394 «9.81 9.98 10.09 01S, 10.2) 10.34 10401046 «10:50 10.53 126 0.00 £00921 9:58 9.77988 995 10.00 101s 10.21 «10.27 | «10.32 10.35 127 0.00 7.67 8.96 936935967 9.78 9.8) 996 10.02 10.09 10.13 10.17 128 0007.33 «8718.13 9.38 9.47955 9.61 8.77984 9.90 9.95 9.99 1290.00 7.00 8458.91 9139.26 935942958 9:58.72 OTRO 30 000-667-821 8.69893 9.05 9.16 9.22 9.40 9k 95S 8H 31000633797 8458S? 868} 993093794897 1 r 132 0.00 6.00 7:73-8:26 852-867-877 HAS LOH 929.20 9:26-9:20 133 0.00 5.67 7.49 -B.05 832 BABS 8.66 BG] «8959.03.99 913. 1 34 0.00533 7.25784 812829 8k] BIBS 8.287 135 0.00 5.007002 7.63-792 810822830852 B BL]. 136 0004.67 679-742-773 791 80412835 «BH BST BH] BOS 137 0.00 4.33656 «-722«758 «7737.86 -795BIB_ «BIBT BAD 138 000° 4.00633 «702-735-735 768777 BOL BIZ BL 828833 139 000 367 610 6a2_717 737751760785 7:98 BOSSI: BIB 140 0.00 3.33 S88 «66398719 7.33 74a 7.69 7.807.907.9702 Kar 000 3.00 S46 643, 80702717727 758 7k 778782787 142 0000-267 $.48 6.2442 BS 700 TMD 37 749759767773 Had 000233 5.23 605 GAS GGG] 54722784 74k 752758 14¢ 0002.00 5.03 «S87 6.27 681.476.787.077. 7.307384 14s 000 67 4.81 5684.10 6.36 GS] -6.63 5.92704 TS 724730 14a 0.00133 4.60550 5.93 G19 635 G47 877 690 TOL 7107.16 14) 0.00 1.00 4.39 «533-5:77 60362) GH2 SU G.75 8796702 148 000 067 4:19-5:15 S60 5.87 ge G17 ws] GTS «GAZE [a9 0.00 0.33 3.99 4.98 5.48 S71 589 G02 6.34 M50 150 000 0.00380 4.81 5285.56 5.78 | 5.876.206.3468 8.55.62 Ls} 0.00 0.00 3.61 a.6s 5:13. S.A] 5.605.736.0620 G33 DAD 1520.00 0.00 342 447497 «5.265.459 5.93607 «62029838 ‘Nites body of eae esis of ox parent delecive cesponding oye vals of quay aden and tangle Sans For Qvale pele ao ea ‘oer. te percent eleive estimate may be read ely om the ube For @ value iss han aco, eae ae mst Be sabe een 100688 ‘SPECIFICATIONS FOR MATERIALS TABLEC. Estimation of Lot Percent Defective by Standard Deviation Method—Conthed Ro Variability Unknown Procedure ‘Standard Deviation Method Quality Enimated Lot Percent Defective for Selected Sample Sizes Index — Q 3 4 5 6 1 8 oo 4s » » 90 153000 000 323 431 48252531 S.aS SSO swe OT GT Ts 000 000 305 415467 4.97517, S.31 S67 5815.95.08 WM 0 ye 136 0.00 000-269-584 4.38 4.69 4.905005 Sak 556 5.70580 1S? 0.00 00) «2:52 369 4.28 4.554.772527Skd SSR 8.88 We) 4X) CC) TH ee Oe PRG CR ye 1S 0.00000 2:19 340 «3:96 4.29 4.51466 SOL 5.20 Sh SAS 1.60 0.00 0002.03, Wo Oke 161 0.00 0.00187 369 4034.25 al AL 497 5123.22 162 0.00 000172 357 391 4.14430 4g? 48G SOL SLT 168 0.00 0001.57 34¢ R79 402 AB ASR 4754.90. 16 0.00 0.00142 331 367 3.90 405 4a? 46479490 1650.00 0.00128 319335 R78 395 436453 Abe 166 0.00 000 LAS 3073433673425 aa 458 8 167 0000.00 Loz 29s 352356373 AIS 432 as 168 0.00 000 «RD 2212s 32D AS 362 AOS. a 16 0.00 000 077 209 «273 «3:10 3343.52 38k 122k 170 0.00 0000.66 «1.98262 2.993283 ae 1B 430 Lm 000 0000357251289 314331378398 aD 420 1720.00 0.0045 1.762.412.7903 3.21 365383399 173 0.00 000036 «165 «230269298 3:1 3563.78 3.90 402 7a eo fee gee Oar sseeke olan ete also) oats sions ge osc 09) 1750.00.00 .19LaS 2249-275 29333735632 3.8 176 0.00 0.00 «O.12—«136 20) «240265283328 347363 3.76 i Ct ey i ee) ee 178 0000000028183 2.22 2.472668 3.30 BT. 790.00 000 «900109 .74= 2.13, 239257 303 3.3383 180 000 0000.00 1.01«1.652.08 2302.49 2933 3.303 Tal 0.00 000 0.00 «093 «1357 196 222 «2402.86 «305322335, [a2 0.00 0.00000 O8S 18) 8S Dt 2322.79 2.98 BRS 327 1830.00 0.00000 078141180206 2.25 2.712.900 307 89 Wo) Oe Te eS i A 185 0.00 0000.00 06s © «1.26165 191209256 2:75-2:92 308 186 0.00 0.00 000037 «19158202248 2.68 28S 297 187 0.00000 0000S! 11251181952] 261278 2.90 1g 0.00 0.00 0.00 «0461081481708 28k 25K 27128. 189 0.00 0.00 0000009913713 8] 22k 247288277 1.9 0.00 000 000 035 093 131 156 4.75 2.21240 2572.70 191 000 0.0 000 «030 «O87 128150682238 251268 192 000-900 0.00026 «ost 118s .6z 208227 24s 257 193 0.00 0.00 0.00 «022-076 «1213756202221 2.38251 19% 0.00 0.00 000018070 1071321501. 232 2S 195 0.60 0.00 0.00 OSS 101126190 2.09225 2.39 196 0.00 0.00 000-012 060096120 138203220233, 197 0.00 0.00 0.0000) ss 9] S133 B97 2S 2.27 198 0.00 0.000.000.0731] tO aT 7392, 2.092 199 000 0000.00 «00S 047 sl 10522167 1.86 20326 200 0.00 0.00 0.00 0.03043 0.78.00 IT 6218198210 201 0.00 0.00 «0.000238 0.72 O9S RST 761932 202 000 000.00 01036067 0.900728 2.00 203 0.00 0.00 0.00000 -032 0.63 ORG Sa? 16618298 20s 0.09 0.00 0.00000 029059 OR? ak LLL] 1.90 205 0.00 0.000.000.0008 08S OT oes TSG TS 18S 1100 624 sar 399 387 375 5.64 32. S41 519 5.08, 498 47 aor 4s oar 438 429 419 410 401 393 3a 3.16 36 359 351 ‘Nomen in body of ele ae eninate of lot prcem defen conespondng to spc valucs of aualiy aden and spl den Foy Q values restr an 6 tbe. the percent defeive imate ay be Fed aie fom the ant For Q values les than seo, he table val me be arate oe 10oc letetintanarn hal elaine aC a nnn rae ASAE Oana ene nN TE Re ee Ro ‘SPECIFICATIONS FOR MATERIALS 689 TABLE C. Estimation of Lot Percent Defective by Standard Deviation Method—Contnued ‘ariablity—Unknown Procedure Standard Deviation Method Gane ____Etinated Lo Percent Defective for Selected Sample Sizes Indes 3 4s 6 7 5 SoM © 2 mf “206 0.00 0.00 “000 00 O27 ost 073 Om a3 ST SRD. 207 9.000.000.0090 0.2) 0.480.700 2R Na? 208 0.00 -0.000.000.00 O18 04s 066 Sk nk 189 207 0.0 0.000.000.0006 04? 062Oe20 8S 16675 240 0.00 0.00000 0.00 ols 039058 od 6 kT 211 0.00 0.00 0.00 0.00012 0.36 08S OT) 1230S 186 212 0.00 0.00 0.00 0.00 oto 033082067 10816? Sk 1 8 213 0.00 0.00 0.00 0.00 0080.30 049s = os 122881501 SB 214 0.00 0.00 0.00 0.00007 ok 0s S110 tsk d6 138 215 0.60 0.00 0.00 0.000.060.2608 SK OT Ae 130,142.50 216 000 0.00 0.00 0.000503] OSS 083 1026. 217 0.900.000.0000 oe 02] osk sz 09007223 218 0.00 0.00 0.00 0.00 003019036 049870319130 1.39 219 060 © 0.00 0.00 -9.00002 07033 oes 0830) S 1271.88 220 0.00 000 000 000 00: 016 031 084 080 097 112 123 13I 221 000 0.00 0.00 0.00 00! aia 029 ol 0.77094 09,120.28 22 000 0.00 0.00 0.000} S13 027039074 = 090 OST 25 223 0.00 0.00 000 0.00 000 «olf 025037 O78? L023 224 0.00 0.00000 0.00 0.00 10023 0se 0808S) LDL 225 0.00 0.00 0.000.000.0000) O21 0320.66 «082096 1071S 225 © 000 0.00 0.90 0.00 0.00 0.08 «020030063 0.79093 LOLI 227 0.00 0.00 0.00 0.000.000.0718 02981078 «0.90 Lor 1.08 228 © 000 0.00 0.00 0.00 0.00 0.08017 G27 S878 ORB 08K 1.06, 229 0.00 0.00 0.00 0.00 0.00 0050S, 0.25056 OT ORS 095103 230° 000 0.00 0.00 0.00 0.000.040.2305 06802093 1.00 231 0.00 0.000.000.0000 0.040322 0520.66 © ORD 0 O97 232 0.00 0.00 0.00 0.00 0.00 0030.11 0.20 049k 077087 O.95 233 9.00 0.00 0.00 0.0 0.00 0.02 a0 01904? O61 075SORS 8D 23¢ 0.00 0.00 0.00 0.00 0.00 002 009 018 04s 059072082090 235 0.00 0.00 0.00 0.00 0.00002 008 ls} 870.70 08D ORT 236 0.00 9.90 0.00 0.000.000.1008 OS. Oa? 05S 088 OBS 237 000 0.90 0.00 0.00 -9.00 GOL 007 ts 0D 0530658 «0750.8? 238 000 0.00 0.00 0.000.000.1005 O13, Osk 05108307300 239 000 0.00 0.0 000 0.00000 .0S, 2049S] ON! O8 240 0400.00 0.00 0.000.000.0005 Ol 0.3S OAT 050069075. 241 0.00 0.00 0.00 0.00 «0.000.000.0808 OS 0570873 242 000 0.00 0.00 0.00 0.00 0.00003. «0.08 «0.300254 0689 242 0.00 0.00 0.00 0.600.000.0003 0.080.300? 054 063.9 244 0.00 0.00 0.00 0.00.00 0.000.030.0827 0S2 OST 245 0.00 0.00 0.00 0.000.000.0002 0.07027 03905005965 248 0.00 0.00 0.0 0.00 0.00 0.00 0020.05 «026 «037 Os 057068 247 0.00 0.00 0.00 0.00 0.00 0.00002 0.050.280.3647 05S 02 248 0.00 0.00 000-000-000 0.00 als. 2k 0344S Oh 0.60 249 000 0.00 0.00 0.00 0.00 0000105023033 M$ «OS? OSB 250 0.00 0.00 0.00 0.00 0.000.000.0104 oat 042 0500.56 251 000 0.000.000 «0.00 0.00 0.0) 0.00.20, oa 0490s 252 000 0.00 0.000.000.0000 Gots. 029° 0470583, 253° 000 0.00 0.00 0.00 «0.000.000.0003. 038046 O51 254 0000.00 0.00 0.00.00 «0.00.00 0.03t? 037 04s 030 255 000 0.00 0.00 «0.00 «9.00 0.00 0.000.027 03504308 256 0.00 0.00 0.000.000.0000 0.00.2, 034 04107 257 0.00 0.00 0.00 © 0.90.00 0.000.000. 033 0400.46 258 0.0 0.00 0.0 © 0.00 «0.000.000.0001 032 039 04s 259 0000.00 0.00 0.00 0.00 0.00.00 ats .22 030038 0.43 ‘bers boy of ble ae esas of ot perce defenva conesponieg wo spac values of quality ida and fmple san” For Qvale peur amo ua ‘wate he pce defective estimate may be Feed dec Fo he ule For Q vals lex han rer, te ble value most be subte fon H0