In most of the statistical quality control experiment, it is not possible to perform 100% inspection due to various reasons. The acceptance sampling plans are concerned with accepting or rejecting a submitted lot of a large size of products on the basis of the quality of the products inspected in a sample taken from the lot. An acceptance sampling plan is a specified plan that establishes the minimum sample size to be used for testing. In most acceptance sampling plans for a truncated life test, the major issue is to determine the sample size from a lot under consideration. Traditionally, when the life test indicates that the mean life of products exceeds the specified one, the lot of products is accepted, otherwise it is rejected. For the purpose of reducing the test time and cost, a truncated life test may be conducted to determine the smallest sample size to ensure a certain mean life time/percentile life time of products when the life test is terminated at a time, t0 and the number of failures observed does not exceed a given acceptance number c.

A common practice in life testing is to terminate the life test by a predetermined time t0 and note the number of failures. One of the objectives of these experiments is to set a lower confidence limit on the mean life/percentile lifetime. It is then to establish a specified mean life with a given probability, which provides protection to consumers. The test may be terminated before the time is reached or when the number of failures exceeds the acceptance number c in which case the decision is to reject the lot.

In the literature of acceptance sampling, there are many methods of testing. Epstein [2] was the first who considered truncated life tests in the exponential distribution. Truncated life tests are considered by many authors for various distributions: for example Sobel and Tischendrof [3], Gupta and Groll [4], Kantam and Rosaiah [5], Baklizi and El Masri [6], Tsai and Wu [7], Balakrishnan et al. [8], Kantam et al. [9], Rao et al. [10] and the references there in.

The above authors considered the design of acceptance sampling plans based on the population mean under a truncated life test. Gupta [11] suggested that for a skewed distribution the median represents a better quality parameter than the mean. On the other hand, for a symmetric distribution, mean is preferable to use as a quality parameter. Lio et al. [12, 13] considered acceptance sampling plans based on the truncated life tests to Birnbaum-Saunders distribution and Burr type XII for percentiles and they proposed that the acceptance sampling plans based on mean may not satisfy the requirement of engineering on the specific percentile of strength or breaking stress. When the quality of a specified low percentile is concerned, the acceptance sampling plans based on the population mean could pass a lot which has the low percentile below the required standard of consumers. Furthermore, a small decrease in the mean with a simultaneous small increase in the variance can result in a significant downward shift in small percentiles of interest. This means that a lot of products could be accepted due to a small decrease in the mean life after inspection. But the material strengths of products are deteriorated significantly and may not meet the consumer’s expectation. Therefore, engineers pay more attention to the percentiles of lifetime than the mean life in life testing applications. Moreover, most of the employed life distributions are not symmetric. Actually, percentiles provide more information regarding a life distribution than the mean life does. When the life distribution is symmetric, the 50^th percentile or the median is equivalent to the mean life. Hence, developing acceptance sampling plans based on percentiles of a life distribution can be treated as a generalization of developing acceptance sampling plans based on the mean life of items. Balakrishnan et al. [8] proposed the acceptance sampling plans could be used for the quantiles and derived the formulae whereas Lio et al. [12, 13] developed for the acceptance sampling plans for any other percentiles of the Birnbaum-Saunders (BS) and Burr Type XII models. They have developed the acceptance sampling plans for percentile by replacing the scale parameter by the 100qth percentile. Rao and Kantam [14] and Rao et al. [15] developed acceptance sampling plans from truncated life tests based on the log-logistic and inverse Rayleigh distributions for Percentiles. Rao et al. [16] developed acceptance sampling plans for percentiles assuming linear failure rate distribution. Balamurali et al. [17] developed acceptance sampling plans based on median life for Fréchet distribution. These reasons motivate to develop acceptance sampling plans based on the percentiles, since odds exponential log logistic distribution (OELLD) is a skewed distribution, we prefer to use the percentile point as the quality parameter, and it will be denoted by. The rest of the paper is organized as follows. In Section 2, we describe concisely the OELLD. The design of proposed acceptance sampling plan for lifetime percentiles under a truncated life test is discussed in Section 3. In Section 4, we present the description of the proposed plan and obtain the necessary results. An example with real data set is also given as an illustration. Finally, conclusions are made in Section 5.

In this section, we provide a brief summary about the odds exponential log logistic distribution. The OELLD was introduced and studied quite extensively by Rosaiah et al. [1]. The probability density function (pdf) and cumulative distribution function (cdf) of OELLD respectively are given as follows:

(1)

(2)

where, λ, σ are the scale parameters and θ respectively shape parameters.

The 100q-th quantile of the OELLD is given as:

(3)

Hence, for the fixed values of λ = λ ₀ and θ = θ ₀ , the quantile t_q given in Equation (3) is the function of scale parameter σ = σ ₀, that is , where

(4)

Note that σ ₀ also depends on λ ₀ = θ ₀, to build up acceptance sampling plans for the OELLD ascertain t_q ≥ t_q ⁰, equivalently that σ exceeds σ ₀.

In this paper, we obtain the minimum sample size necessary to ensure a percentile life when the life test is terminated at a pre-assigned time t ₀^q and when the observed number of failures does not exceed a given acceptance number c. The decision procedure is to accept a lot only if the specified percentile of lifetime is established with a pre-assigned high probability α, which provides protection to the consumer. The life test experiment gets terminated at the time at which (c + 1)^th failure is observed or at quantile time t_q, whichever is earlier. Acceptance sampling plan based on truncated life tests for the OELLD using average life was discussed by Rosaiah et al. [1].

The probability of accepting lot based on the number of failures from all groups under a truncated life test at the test time schedule is given by

(5)

Where n is the sample size, c is the acceptance number, and p is the probability of getting a failure within the life test schedule t ₀,. If the product lifetime follows an OELLD, then p = F(t ₀). Usually, it would be convenient to determine the experiment termination time t ₀ as t ₀ = δ_q ⁰ t_q ⁰ for a constant δ_q ⁰ and the targeted 100q-th lifetime percentile, t _q ⁰. Suppose t_q is the true 100q-th lifetime percentile. Then, p can be rewritten as

(6)

In order to find the design parameters of the proposed acceptance sampling plan, we prefer the approach based on two points on the OC curve by considering the producer’s and consumer’s risks. In our approach, the quality level is measured through the ratio of its percentile lifetime to the lifetime, t_q/t_q0. These percentile ratios are very helpful for the producer to enhance the quality of products. From the producer’s perspective, the probability of lot acceptance should be at least 1 - α at the acceptable reliability level (ARL), p₁. So the producer demands that a lot should be accepted at various levels, say t_q/t_q0 = 2, 4, 6, 8, 10 in Equation (5). On the other hand, from the consumer’s viewpoint, the lot rejection probability should be at most β at the lot tolerance reliability level (LTRL), p₂. In this way, the consumer considers that a lot should be rejected when t_q/t_q⁰ = 1. From Equation (5), we have

(7)

(8)

where p₁ and p₂ are given by

(9)

The plan parametric quantities for distinct values of parameters λ are θ constructed. Given the producer’s risk α = 0.5 and termination time schedule t ₀ = δ_q ⁰t_q ⁰ with δ_q ⁰ = 1.0,1.5,2.5,2.0,3.0, the two parameters of the proposed sampling plan under the truncated life test at the pre-specified time, t ₀, with λ = 2 and θ = 0.1,1.0,2.0 are obtained according to the consumer’s confidence levels β = 0.25,0.10,0.05,0.01 for 50^th percentile and the operating characteristic values are also calculated and the results are presented in Tables 1-3. The plan parameters are presented in Tables 1-3 for λ = 2 and θ = 1.0,1.5,2.0 with 50^th percentiles, whereas Table 4 shows the plan parameters for and are the MLE’s from the data given in Subsection 4.2 at 50^th percentile. We noticed from Tables 1-4 that the percentile ratio increases, the sample size n decreases for all the cases. The sampling plans for OELLD were developed by Rosaiah et al. [1].

In this article, we developed the single acceptance sampling plans based on the OELLD percentiles, when the life test is truncated at a pre-fixed time. To ensure that the life quality of products exceeds a specified one in terms of the percentile lifetime, the acceptance sampling plans based on percentiles can be used. We have designed sampling plans based on two-points on the OC curve approach for assuring percentile lifetime of the products. Some tables are provided for practical use in industry and also proposed plan illustrated with real data set. We fitted the proposed OELLD curve for the above data which is shown in Fig. (1).