Service level

Last updated July 31, 2024

Service level measures the performance of a system, service or supply. Certain goals are defined and the service level gives the percentage to which those goals should be achieved.

Percentage of calls answered in a call center
Percentage of customers waiting less than a given fixed time
Percentage of customers that do not experience a stockout
Percentage of all parts of an order being fulfilled completely
Use of a safety stock to ensure that a target percentage of orders can be met in full and on time.^[1]

The term "service level" is used in supply-chain management and in inventory management to measure the performance of inventory replenishment policies.^[1] Under consideration, from the optimal solution of such a model also the optimal size of back orders can be derived. A back order is an order placed for an item which is out-of-stock and awaiting fulfillment.^[2] Unfortunately, this optimization approach requires that the planner knows the optimal value of the back order costs. As these costs are difficult to quantify in practice, the logistical performance of an inventory node in a supply network is measured with the help of technical performance measures. The target values of these measures are set by the decision maker.

Definitions and typology

Several definitions of service levels are used in the literature as well as in practice. These may differ not only with respect to their scope and to the number of considered products but also with respect to the time interval they are related to. These performance measures are the key performance indicators (KPI) of an inventory node which must be regularly monitored. If the controlling of the performance of an inventory node is neglected, the decision maker will not be able to optimize the processes within a supply chain.

α service level (type 1)

The α service level is an event-oriented performance criterion. It measures the probability that all customer orders arriving within a given time interval will be completely delivered from stock on hand, i.e. without delay.

Two versions are discussed in the literature differing with respect to the time interval within which the customers arrive. With reference to a demand period, α denotes the probability that an arbitrarily arriving customer order will be completely served from stock on hand, i.e. without an inventory-related waiting time (period $\alpha _{p}$ service level):

$\alpha _{p}=Prob\{Period~demand\leq \;{Inventory~on~hand~at~the~beginning~of~a~period}\}$ .

In order to determine the safety stock that guarantees a target $\alpha _{p}$ service level, the stationary probability distribution of the inventory on hand must be known. This version of α is also called the ready rate.

If an order cycle is considered as the standard period of reference, then α denotes the probability of no stockout within an order cycle which is equal to the proportion of all order cycles with no stockouts (cycle $\alpha _{c}$ service level):

$\alpha _{c}=Prob\{Demand~during~replenishment~lead~time\leq Inventory~on~hand~at~the~beginning~of~the~lead~time\}$

This second definition, which is often used in operations management textbooks, is based on the idea of not running out of stock during the time between re-ordering and order arrival (the leadtime). That is, the probability of demand during that leadtime being less than or equal to the amount of stock you had left when you ordered. It assumes your reorder point is positive, that orders are in unit increments and inventory is monitored continuously so you cannot stock out prior to reordering.

β service level (type 2)

The β service level is a quantity-oriented performance measure describing the proportion of total demand within a reference period which is delivered without delay from stock on hand:

$\beta =1-{\frac {Expected~backorders~per~time~period}{Expected~period~demand}}$

This is equal to the probability that an arbitrary demand unit is delivered without delay. This approach usually involves calculating a loss integral, whose values are tabulated for the normal distribution.^[3]

Because, contrary to the variations of the $\alpha$ service level, the $\beta$ service level does not only reflect the stockout event but also the amount backordered, it is widely used in industrial practice.

Also, by the definitions, comparing service levels we have $\alpha \leq \beta$ whenever the probability of zero demand equals 0.

γ service level

The γ service level, a time- and quantity-related performance criterion, serves to reflect not only the amount of backorders but also the waiting times of the demands backordered. The γ service level is defined as follows:

$\gamma =1-{\frac {Expected~backorder~level~per~time~period}{Expected~period~demand}}$

The γ service level is rarely used in industrial practice.

Service rate

In business, service rate is a performance metric used to measure the customer service in a supply organization. One example of a service rate measures the number of units filled as a percentage of the total ordered and is known as fill rate. Fill rate is different from service level. If a customer orders 1000 units, and their supplier can only provide 900 units of that order, their fill rate is 90%.
In statistics, notably in queuing theory, service rate denotes the rate at which customers are being served in a system. It is the reciprocal of the service time. For example, a supermarket cash desk with an average service time of 30 seconds per customer would have an average service rate of 2 per minute. In statistics the Greek letter $\mu$ is used for the service rate.

Terminology

The term "Service Level Agreement" (SLA) is frequently used for all aspects of a service level, but in more precise use one may distinguish:^[4]

Service Level Indicator (SLI): measures of service level, like availability (uptime);
Service Level Objective (SLO): objectives based on these indicators, like 99.95% availability;
Service Level Agreement (SLA): contract based on these objectives; a sample clause may be "if availability is 99% to 99.95% in a given month, the customer gets 10% off their monthly bill".^[5]

SLIs form the basis of SLOs, which in turn form the basis of SLAs. If an SLO is missed, customers will typically receive a credit or rebate, as stipulated by the SLA. A missed SLO is sometimes casually referred to as an SLA violation, but this is actually within the scope of the SLA; if an SLA itself is violated (e.g., by not giving a rebate for a missed SLO), it is instead likely to result in a court case for breach of contract.^[4]

Related Research Articles

In probability theory and statistics, the exponential distribution or negative exponential distribution is the probability distribution of the distance between events in a Poisson point process, i.e., a process in which events occur continuously and independently at a constant average rate; the distance parameter could be any meaningful mono-dimensional measure of the process, such as time between production errors, or length along a roll of fabric in the weaving manufacturing process. It is a particular case of the gamma distribution. It is the continuous analogue of the geometric distribution, and it has the key property of being memoryless. In addition to being used for the analysis of Poisson point processes it is found in various other contexts.

The Pareto distribution, named after the Italian civil engineer, economist, and sociologist Vilfredo Pareto, is a power-law probability distribution that is used in description of social, quality control, scientific, geophysical, actuarial, and many other types of observable phenomena; the principle originally applied to describing the distribution of wealth in a society, fitting the trend that a large portion of wealth is held by a small fraction of the population. The Pareto principle or "80-20 rule" stating that 80% of outcomes are due to 20% of causes was named in honour of Pareto, but the concepts are distinct, and only Pareto distributions with shape value of log₄5 ≈ 1.16 precisely reflect it. Empirical observation has shown that this 80-20 distribution fits a wide range of cases, including natural phenomena and human activities.

Growth accounting is a procedure used in economics to measure the contribution of different factors to economic growth and to indirectly compute the rate of technological progress, measured as a residual, in an economy. Growth accounting decomposes the growth rate of an economy's total output into that which is due to increases in the contributing amount of the factors used—usually the increase in the amount of capital and labor—and that which cannot be accounted for by observable changes in factor utilization. The unexplained part of growth in GDP is then taken to represent increases in productivity or a measure of broadly defined technological progress.

Empirical Bayes methods are procedures for statistical inference in which the prior probability distribution is estimated from the data. This approach stands in contrast to standard Bayesian methods, for which the prior distribution is fixed before any data are observed. Despite this difference in perspective, empirical Bayes may be viewed as an approximation to a fully Bayesian treatment of a hierarchical model wherein the parameters at the highest level of the hierarchy are set to their most likely values, instead of being integrated out. Empirical Bayes, also known as maximum marginal likelihood, represents a convenient approach for setting hyperparameters, but has been mostly supplanted by fully Bayesian hierarchical analyses since the 2000s with the increasing availability of well-performing computation techniques. It is still commonly used, however, for variational methods in Deep Learning, such as variational autoencoders, where latent variable spaces are high-dimensional.

A service-level agreement (SLA) is an agreement between a service provider and a customer. Particular aspects of the service – quality, availability, responsibilities – are agreed between the service provider and the service user. The most common component of an SLA is that the services should be provided to the customer as agreed upon in the contract. As an example, Internet service providers and telcos will commonly include service level agreements within the terms of their contracts with customers to define the level(s) of service being sold in plain language terms. In this case, the SLA will typically have a technical definition of mean time between failures (MTBF), mean time to repair or mean time to recovery (MTTR); identifying which party is responsible for reporting faults or paying fees; responsibility for various data rates; throughput; jitter; or similar measurable details.

Economic order quantity (EOQ), also known as financial purchase quantity or economic buying quantity, is the order quantity that minimizes the total holding costs and ordering costs in inventory management. It is one of the oldest classical production scheduling models. The model was developed by Ford W. Harris in 1913, but the consultant R. H. Wilson applied it extensively, and he and K. Andler are given credit for their in-depth analysis.

Cohen's kappa coefficient is a statistic that is used to measure inter-rater reliability for qualitative (categorical) items. It is generally thought to be a more robust measure than simple percent agreement calculation, as κ takes into account the possibility of the agreement occurring by chance. There is controversy surrounding Cohen's kappa due to the difficulty in interpreting indices of agreement. Some researchers have suggested that it is conceptually simpler to evaluate disagreement between items.

In information theory, the Rényi entropy is a quantity that generalizes various notions of entropy, including Hartley entropy, Shannon entropy, collision entropy, and min-entropy. The Rényi entropy is named after Alfréd Rényi, who looked for the most general way to quantify information while preserving additivity for independent events. In the context of fractal dimension estimation, the Rényi entropy forms the basis of the concept of generalized dimensions.

Safety stock is a term used by logisticians to describe a level of extra stock that is maintained to mitigate risk of stockouts caused by uncertainties in supply and demand. Adequate safety stock levels permit business operations to proceed according to their plans. Safety stock is held when uncertainty exists in demand, supply, or manufacturing yield, and serves as an insurance against stockouts.

The relative risk (RR) or risk ratio is the ratio of the probability of an outcome in an exposed group to the probability of an outcome in an unexposed group. Together with risk difference and odds ratio, relative risk measures the association between the exposure and the outcome.

Nuclear reactor physics is the field of physics that studies and deals with the applied study and engineering applications of chain reaction to induce a controlled rate of fission in a nuclear reactor for the production of energy. Most nuclear reactors use a chain reaction to induce a controlled rate of nuclear fission in fissile material, releasing both energy and free neutrons. A reactor consists of an assembly of nuclear fuel, usually surrounded by a neutron moderator such as regular water, heavy water, graphite, or zirconium hydride, and fitted with mechanisms such as control rods which control the rate of the reaction.

In queueing theory, a discipline within the mathematical theory of probability, a Jackson network is a class of queueing network where the equilibrium distribution is particularly simple to compute as the network has a product-form solution. It was the first significant development in the theory of networks of queues, and generalising and applying the ideas of the theorem to search for similar product-form solutions in other networks has been the subject of much research, including ideas used in the development of the Internet. The networks were first identified by James R. Jackson and his paper was re-printed in the journal Management Science’s ‘Ten Most Influential Titles of Management Sciences First Fifty Years.’

The reorder point (ROP), also reorder level (ROL) or "optimal re-order level", is the level of inventory which triggers an action to replenish that particular inventory. It is a minimum amount of an item which a firm holds in stock, such that, when stock falls to this amount, the item must be reordered. It is normally calculated as the forecast usage during the replenishment lead time plus safety stock. In the EOQ model, it was assumed that there is no time lag between ordering and receipt of materials.

<span class="mw-page-title-main">Precision and recall</span> Pattern-recognition performance metrics

In pattern recognition, information retrieval, object detection and classification, precision and recall are performance metrics that apply to data retrieved from a collection, corpus or sample space.

Material theory is the sub-specialty within operations research and operations management that is concerned with the design of production/inventory systems to minimize costs: it studies the decisions faced by firms and the military in connection with manufacturing, warehousing, supply chains, spare part allocation and so on and provides the mathematical foundation for logistics. The inventory control problem is the problem faced by a firm that must decide how much to order in each time period to meet demand for its products. The problem can be modeled using mathematical techniques of optimal control, dynamic programming and network optimization. The study of such models is part of inventory theory.

In probability theory and statistics, the Poisson distribution is a discrete probability distribution that expresses the probability of a given number of events occurring in a fixed interval of time if these events occur with a known constant mean rate and independently of the time since the last event. It can also be used for the number of events in other types of intervals than time, and in dimension greater than 1.

In queueing theory, a discipline within the mathematical theory of probability, a fluid queue is a mathematical model used to describe the fluid level in a reservoir subject to randomly determined periods of filling and emptying. The term dam theory was used in earlier literature for these models. The model has been used to approximate discrete models, model the spread of wildfires, in ruin theory and to model high speed data networks. The model applies the leaky bucket algorithm to a stochastic source.

The public market equivalent (PME) is a collection of performance measures developed to assess private equity funds and to overcome the limitations of the internal rate of return and multiple on invested capital measurements. While the calculations differ, they all attempt to measure the return from deploying a private equity fund's cash flows into a stock market index.

The base stock model is a statistical model in inventory theory. In this model inventory is refilled one unit at a time and demand is random. If there is only one replenishment, then the problem can be solved with the newsvendor model.

The (Q,r) model is a class of models in inventory theory. A general (Q,r) model can be extended from both the EOQ model and the base stock model

References

1 2 Aberdeen Group, Are Your Inventory Management Practices Outdated?, published on 1 March 2005, accessed on 14 July 2024
↑ Kenton, W., Backorder: Definition, Causes, Example, Vs. Out-of-Stock, updated on 26 December 2022, accessed on 9 February 2024
↑ Donald Bowersox, David Closs, M. Bixby Cooper, Supply Chain Logistics Management, McGraw-Hill 2012
1 2 Niall Richard Murphy; Betsy Beyer; Jennifer Petoff; Chris Jones. "Service Level Terminology". Site Reliability Engineering: How Google Runs Production Systems. pp. 37–40.
↑ For example, "Google Compute Engine Service Level Agreement (SLA)" . Retrieved 2016-05-30.