WarningThis information has been archived for reference or research purposes.

Archived Content

Information identified as archived on the Web is for reference, research or recordkeeping purposes. It has not been altered or updated after the date of archiving. Web pages that are archived on the Web are not subject to the Government of Canada Web Standards. As per the Communications Policy of the Government of Canada, you can request alternate formats on the "Contact Us" page.


Live exercises

DND Photo IS2002-6531a by Master Corporal Paul MacGregor

A battery Sergeant-Major watches an M-109 howitzer firing during a live fire exercise at CFB Petawawa.

Approaches to Determining Army Operational Stockpile Levels

by William S. Andrews and William J. Hurley

Print PDF

For more information on accessing this file, please visit our help page.

Planning for the operational deployment of Canadian Forces (CF) units and formations offshore involves consideration of a number of factors, including the amount of all types of ammunition likely to be required. As these items are rarely ‘off the shelf’, stockpile levels must be established and maintained in anticipation of such deployments. Requisite ammunition levels, or scales, have been in existence for decades, although they have largely been based on Second World War and Korean War data and in anticipation of war with the Warsaw Pact forces in Europe. The world, however, has changed.

Current Canadian policy statements and planning guides are published annually by the CF. They delineate the various responsibilities for stockpiles, including ammunition, in support of operations.1 For example, the Chief of the Land Staff (CLS) is “responsible for ensuring that stocks are available in sufficient quantities”. The Deputy Chief of the Defence Staff (DCDS) is responsible “for setting stock levels and assuring the availability of operational level sustainment stocks”, and the Assistant Deputy Minister (Materiel) [ADM (Mat)] for “setting stock levels and assuring the availability of strategic level sustainment stocks”. The stock levels are to be based on sustaining the Vanguard of the Main Contingency Force “indefinitely in a low threat environment” and the Main Contingency Force “for 60 days of combat operations”.2

This paper discusses and assesses methodologies for determining ammunition stockpile levels currently available to CF planners, and makes recommendations for methodologies for future use.


Faced with funding reductions of 10 percent imposed as a result of the federal government economic statement on 2 December 1992, the Department of National Defence (DND) conducted an examination of ammunition requirements using a zero-based approach.3 The final report recommended that operational stockpile levels be set at 60 days (30 days basic and 30 days sustainment, or 30/30) for mid-intensity operations. The value of 60 was a compromise in that it considered existing stock levels, the money likely to be available for sustaining longer periods of deployment, and available storage facilities. The report noted that both the Falkland and Gulf wars were of limited duration, so the shorter sustainment periods should be bearable. The difficulty in activating domestic industry and acquiring resources off-shore were noted, and the risk was felt to be acceptable.4

The report referred to above (Note 4) does not mention a methodology used for determining the composition of stock levels, although overall cost values are included. It was mentioned informally to the authors that previous stock levels, perhaps dating back to the Second World War or the Korean War, were simply adopted. The Operational Research and Analysis Branch of DND conducted studies of ammunition expenditure rates in low- to mid- intensity operations in order to provide a “rough validation of the current land force ammunition scales”.5

Van Horn’s A Validation of Canadian Forces Land Ammunition Scales (See Note 5) provides a good discussion of the derivation of these Scales and a comparison of these with Supreme Headquarters Allied Powers Europe (SHAPE), American and British approaches to stock levels. The first and only CF Scale considering all pertinent parameters was issued in 1965. Only one value per munition was quoted, however, implying that only one level of intensity of operations was considered. Further, the values of the Scale were on a per-weapon system basis (e.g., rounds per gun) and covered a period of 30 days, which included both basic (from 3 to 10 days, depending on the munition) and maintenance loads. The values were based on NATO maintenance rates and activity factors for mid-level operations in mid-level conflicts in Europe. Their applicability for other intensities, levels of operations or geographic locations is arguable.

The Canadian approach, however, is consistent with SHAPE Stockpile Planning Guidance (SPG), which considers the first 30-day period of conflict with a Warsaw Pact enemy in Central Europe. It too is felt to be based on Second War and Korean War data, with the January 1990 edition bearing little change from previous versions. The breakdown is slightly different, in that it separates the requirement into two portions: sustainment of the first seven days of combat, followed by the remaining 23 days. The methodology for determining the Stockpile Planning Guidance values was not available.

The American approach, as described in the US Army’s Staff Officers’ Field Manual for Organizational, Technical and Logistical Data Planning Factors of 1965, provides daily consumption rates for all types of munitions for various levels of operations, as well as basic loads for engagements up to 15 days in both the attack and defence. Beyond 15 days, another document, last updated in 1977, is to be used. The basis of these two documents still appears to be Second World War and Korean War data.6 Informal discussions revealed that the US Army is grappling with the stockpile problem as well and has not yet arrived at a suitable solution.

The British approach is to use different Scales for different theatres, with the SHAPE Stockpile Planning Guidance being applicable for Europe. Non NATO/European operations have Scales based on operations in Borneo and South Arabia, and provide daily ammunition expenditure rates (DAERs) for both the first five days and then the balance to 30 days.

The Vincent, Willner and Van Horn report cited (see Note 5) is the second part of the 1992-mandated study of ammunition stock levels. It examined usage rates (levels of effort) employed by Australian, US and British planners for generic munitions, and in some cases made extrapolations to Canadian munitions, for mid, low mid and low levels of operations. Due to the limited time frame available for the study, the report is a self-confessed “quick and dirty assessment of the factors that can be applied to various weapon systems”. In fact, not much information of substance was forthcoming from Allied armies, in part due to time limitations and in part to national security concerns. The rationale for the US Marine Corps values was not provided, but the stock levels quoted by both the Australians and British were level-of-effort or consumption based, as opposed to target based. These data were used, in part, for determining 30-day operational stocks for a few selected Canadian munitions, quoted on a per-weapon basis (rather than on a per-unit basis, as quoted in the report).

Top of Page

The Ammunition Planning Working Group (APWG) was established to review stockpiling criteria and to determine the validity of the Defence Management Committee-imposed 10 percent reduction to ammunition funding, commencing in Fiscal Year 96/97. APWG met fairly regularly. The tenth meeting was conducted on 9 February 19957 with the next meeting occurring on 20 April 1995.8 The February 1995 meeting included a briefing by the Chief of the Land Staff, indicating that he “had accepted the results of the Land Forces opstock [operational stock] review” and that Land Forces Command was reviewing the number of Defence Development Plan 1993 (DDP 93) missions and tasks which could be carried out simultaneously. This latter exercise could result in a reduction in the number of tasks, with a concomitant reduction in ammunition stock requirements. The February 1995 APWG meeting also indicated that a 30/30 sustainment option was likely to be adopted.

The Working Group’s Zero-Based Ammunition and Sonobuoy Review Report to Defence Management Committee9 was completed on 15 May 1995. It included both the Operational Stock Review and the Steady State Review, the latter to determine the annual ammunition requirement. The review was effected by developing and adopting a revised ammunition-planning model based on reduced global threat levels and a move to an emphasis on contingency operations. The model included a steady state training component, a transition component and operational stocks. The operational stock review team included operational research representatives, who provided a scientific approach to stock determination, as well as an independent review of the results and the proposal of alternate methodologies. The other component of the review team involved members of the environmental services (Navy, Army and Air Force), who provided operational judgement and experience. The parameters used to determine the operational stock requirements were the Defence Development Plan 93 force structures, missions and tasks (including modifications in the 1994 White Paper), the 1993 readiness and sustainment policy (up to 120 days), enemy attrition as a result of continuing operations, and operations of mid-level intensity within a coalition.

105mm shell

DND Photo IVD2004-0224 by Sergeant David Snashall

A member of Vancouver’s 15th Field Regiment preparing to load a 105mm shell during a 39th Brigade exercise at the US Army training area at Yakima, Washington.

The Operational Stock Review Team found insufficient empirical or historical data available to estimate ammunition expenditures for all natures within the postulated scenarios. Consequently, the team “relied heavily on military insight and professional judgement” for estimating expenditures. A level of effort methodology (LEM) approach was used to estimate expenditures, based on activity rates, without requiring a detailed knowledge or specification of the threat. Three-point estimates (minimum, most likely and maximum expenditure rates) were made “to capture the uncertainty of future operations by specifying the range of feasible expenditure rates” and also “to define the possible duration of the combat phases to reflect the expected ebb and flow of combat operations”.

For the Army, the three-point methodology was abandoned, due to the complexity of the scenarios and the variety of munitions (over 100 different natures) involved. Instead, they started with the Canadian Land Force Staff Ammunition Scales and used expert judgement to modify these values based on the limiting parameters. Two subtleties were introduced. For direct fire weapons, the combat lifetime approach was adopted. For indirect fire weapons, a level of effort methodology approach based on the number of rounds per fire mission per gun per day was used.

Target-oriented Methodology

A target-oriented methodology (TOM) involves determining the quantity of munitions required to defeat the enemy targets available. It is usually probabilistic, invoking single-shot kill probabilities for given munitions against given targets.

ACROSS. For NATO operations, the NATO Bi-Strategic Commands Stockpile Planning Guidance document (the most recent version being SPG-2001, dated 9 May 2001) provides methodologies embedded in a software programme for determining the appropriate stockpile levels for all munitions.10 SPG-2001 states that stocks for “battle decisive munitions (e.g. missiles, torpedoes) will be calculated utilizing a ‘target oriented’ approach. All other munitions requirements (e.g. ... small arms ammunition) ... will be calculated using a ‘level of effort’ approach.” The software programme for determining these levels is the ACE Resource Optimization Software System (ACROSS), while the appropriate Army portion is the Land Forces Equipment and Munitions Expenditure Module (LEMEM).11

Interestingly, ACROSS introduces a new approach to SPG, a target-oriented methodology for “battle decisive munitions” such as field artillery gun and howitzer rounds, anti-tank gun rounds, cannon rounds of 20mm and above, mortar rounds and tank ammunition. Not included in this group are mines, small arms ammunition, grenades and light short-range anti-tank weapons. The target- oriented methodology includes such factors as lists of enemy targets and own forces holdings, as well as operational parameters. This methodology is time-independent, and applies to missions of both short and long duration, with stocks expressed in units of individual rounds.

Top of Page

The more traditional level of effort methodology is also used in ACROSS, and considers such factors as average consumption per day, number of consumers, number of days and intensity, with the stock levels expressed in days of supply. This methodology would be used for small arms ammunition.


DND Photo IS2002-6533a by Master Corporal Paul MacGregor

An artillery soldier packs up a mortar following the end of Exercise “Rapid Bear” at CFB Petawawa, August 2002.

ACROSS is based on a mathematical programming approach which optimizes one of two objective functions, either total cost or the amount of target value destroyed. In the case where the objective is total cost, the optimal solution specifies an assignment of munitions to targets that destroys all targets at minimum cost. Alternatively, it can assign targets to munitions to maximize target destruction, subject to bounded total costs, while considering target and shooter interdependencies, stowage requirements, and lost equipment costs. These mathematical programmes are relatively small. One example considered 1489 variables and 1263 constraints. ACROSS comes with an extensive database, but users must still provide a significant amount of idiosyncratic force-specific data, including the parameters of their own assets and the costs of individual munitions. Where munitions are unique to a particular nation, effectiveness values against the complete array of targets available must also be provided. Anecdotally, users have remarked on the complexity and extensive user inputs involved in running ACROSS.

ACROSS is not without other weaknesses. The latest version, 3.0, included some 30 software fixes. Taylor’s A Case Against Linear Programming for Stockpile Planning discusses peculiarities of the ACROSS solution for a Canadian case, where provably suboptimal solutions were found by the programme.12 Another chronic weakness is in the treatment of indirect fire support.13 ACROSS uses the target-oriented methodology, but, in actual practice, indirect fire is used against area targets rather than point targets. It does not achieve 100 percent kill, and its effectiveness is degraded by protection. Further, ACROSS currently contains a risk of double counting point targets, which may be attacked and destroyed by one or both of direct or indirect fire weapons. Put differently, the current treatment of indirect fire munitions by ACROSS is at best problematic.

The NATO C3 Agency, responsible for developing ACROSS, is currently addressing the indirect fire challenge by retaining the target-oriented methodology approach but by treating appropriate targets as ‘surface’ as distinct from point. Further target qualifications include protected versus unprotected and the target effects as destroy versus neutralize. Another subtlety is the distinction between close combat and deep targets, with the latter being beyond mortar range. A final recommendation of the working group examining the problem is to remove illumination and smoke rounds, as they are considered “non battle decisive”. Consequently, these munitions and their related fire support missions (illuminating, obscuring and screening) are not considered by ACROSS. This leaves only interdiction and final protective fire missions (but not suppression), for which the quantities of munitions required have yet to be finalized, although UK rates are likely to be adopted.

It is worth noting that the target-oriented approach does not consider fire missions required to neutralize terrain features (speculative fire), for registration or for training. These tasks should all, however, be included in a level of effort approach (although for munitions not covered by target-oriented methodology, it was felt that daily expenditure rates for some systems in ACROSS were unrealistic and “should reflect target density and technical weapons specifications in combination with common military doctrine”.14

MASPP. The Marginal Analysis Stockpile Planning Program (MASPP) was developed by J4 Logistics Analysis at NDHQ in Ottawa. It is also a target-oriented methodology approach and incrementally increases the starting inventory of rounds to defeat a given target list. The confidence level of the solution is determined, so that the analysis proceeds incrementally from a low cost (low inventory), low confidence level to a high cost, high confidence level. The objective function can be a total cost or a specified confidence level (probability of mission success).

MASPP requires a target list and, like ACROSS, an extensive database of single shot kill probabilities (SSKPs) for each munition against each likely target, although these values can be modified by target hardening and ammunition degradation. Single shot kill probabilities data are not always readily available, but are essential for the analysis to succeed. MASPP was evaluated by the British Centre for Defence Analysis (CDA) and found wanting in a number of areas. They felt MASPP lacked tactical awareness, in that there were “serious concerns about the user’s ability to generate the potentially large quantity of target-matching data now required”. The Centre for Defence Analysis was also concerned about the difficulty in generating credible and valid target hardening, ammunition degradation, ammunition to target (reflecting tactics and doctrine by providing the proportion of each target type to be attacked using each munition type), and ammunition usage (reflecting the proportional allocation of munitions to scenarios) data.

Top of Page

In comparison with ACROSS, the Centre for Defence Analysis found that MASPP was generally less sophisticated (missing an internal database, including weapon system and platform data), did not consider time and space, shooter attrition or target interdependencies and was lacking in output data analysis.15

In addition to the Centre for Defence Analysis’ concerns, neither ACROSS nor MASPP addresses speculative fire by direct and indirect fire systems, rounds used in training in theatre, logistic loss and model risk. Perhaps most significant, there does not seem to be any explicit recognition of breakpoints or engagement termination data, as discussed in Helmbold’s Decision in Battle16, where historical data suggest that a combatant can be considered defeated after suffering some 30 percent losses. (This could be addressed easily within MASPP, though, by simply adjusting the target set.) Addressing breakpoints is consistent with Joint Munitions Effectiveness Manuals, which set the value at 30 percent.17 Put differently, it does not make sense to plan on the basis of defeating all the targets in a target set.

Examples of the application of MASPP to specific or generic munitions can be found in Notes 18 to 22.

Level of Effort Methodology

This methodology involves considering historical or simulation (war gaming) data to determine usage or consumption rates for given weapon systems. It is not dependent upon specific scenarios and thus can be more generic than target-oriented methodology. The weakness or challenge comes in using data that may lack fidelity with proposed deployments. In fact, level of effort methodology has been the traditional approach to determining stockpile levels, with the datum being Second World War and/or Korean War consumption rates. Without simulation, the database may be somewhat limited and probably not statistically significant. Simulation can be used to augment empirical and historical data, but the ability of gamers to consider the variety of modes of expenditure (direct and indirect fire, suppression, speculative fire, training and stock losses) is crucial and may well be beyond the capabilities of many, if not most, players.

The level of effort method is fairly robust in that often, for indirect fire munitions in particular, expenditure levels are governed by such in-theatre factors as transportation, dumping, resupply, and the need for the firing platforms to redeploy due to air and counter battery attacks. Other factors, such as the ability to deliver munitions into theatre and even to regenerate stocks from the domestic industrial base, also impose limitations in use. All these factors are included in historical usage figures, but the circumstances may have changed significantly from when the data were generated.

Examples of the application of level of effort methodology to problems are contained in Notes 23 and 24. For comparison, the application of expert opinion separate from either target-oriented or level of effort approaches is found in Gauthier’s Air Force Operational Stockpile Requirements.25

Allied Approaches

Great Britain. As noted above, the British are determining daily ammunition expenditure rates based on NATO or historical level of effort methodology data, as appropriate. All this is used within Waymarker, which is described as not a model as such, but rather a “roadmap” to ensure that a logical sequence is followed for determining and maintaining stockpile levels. Actual munition levels would still be determined by any or all of history, modeling or judgement.26

United States. Stockpile determination is currently being addressed by the American military, as well. The Department of Defense Instruction Number 3000.4 (Capabilities-Based Munitions Requirements Process27) outlines the responsibilities of various offices for determining and establishing munitions requirements, without specifying any particular methodology. It does allow planners to base munitions requirements on two factors: the force structure assigned to a mission and the enemy or “threat” to be defeated. The latter factor suggests a target- oriented approach. Combined with these combat requirements (based on the phased threat distribution) are the strategic readiness requirement (to arm forces not committed to combat operations), and current operation and forward presence requirements for forces deployed but not committed to particular threats. All these combined factors constitute the war reserve munitions requirements. When the training and testing requirement is added, the sum forms the total munitions requirement. This total is then compared to the current inventory to determine the amount outstanding. At this point, other factors, such as funding, are applied to determine the extent to which the total munitions requirement will be filled.

The US Army is investigating a number of logistics concepts aimed at achieving a transition from a stockage-based, decentralized system to a centrally managed, distribution-based system fully synchronized with velocity management and total asset visibility. “The Velocity Management program will provide improved material support through aggressive application of systematic defining, measuring, and improving business practices. ... Automated systems will provide ... tools that will analyze the stocks necessary to meet the requirements of deployment and garrison forces, including split-based operations ... with a streamlined support base, integrating vendor and military supply pipelines that are more aligned with just-in-time logistics. ... Stockage levels will be reduced to items that support the customer’s immediate needs and readiness requirements”.28 This, in effect, appears to be an application of commercial delivery and stock management practices to military requirements. Again, specific methodologies are not discussed.

Top of Page


Two methodologies have been discussed above as tools for determining stockpile levels: target-oriented and level of effort. The latter is the traditional approach in use in Canada for decades. It is a straightforward approach for many munition types. However it is problematic for munitions that have unique capabilities. In the past, this has been addressed by “expert opinion”, without any apparent attempt to validate these opinions. With the introduction of ACROSS and MASPP, target-oriented methodologies are now available. Although appearing deterministic at first glance, and thus more reactive to new munitions and scenarios, they are in fact heavily dependent upon user- supplied data on target effectiveness and costs, while ignoring situational awareness (informed opinions on how to destroy specific targets with specific munitions). They are not able to directly address expenditures due to speculative and suppressive fire, in-theatre training, logistic loss, and model risk. Further, employing optimization routines, they may not adequately reflect the vagaries of combat, where munitions are not necessarily used against the targets for which they were designed, nor to the precision achieved outside the “fog of war”. Finally, they are highly dependent upon a reasonably accurate (quantitative) target assessment, while ignoring the concept of a breakpoint, or level of loss, beyond which defeat can reasonably be assumed. In short, there are significant weaknesses with both approaches.

Gun crew

DND Photo IVD2004-0229 by Sergeant David Snashall

A 15th Field Regiment gun crew firing a 105mm howitzer during a Militia Brigade exercise at Yakima, Washington, March 2004.

To address these weaknesses, a decision process termed the Operational Stock Planning Process (OSPP) is proposed below. There are a number of factors, though, that are not mentioned explicitly in the model description, but are felt to warrant explicit, albeit brief, discussion.

Changing Guidelines. A new White Paper, or, at least, a defence policy statement, is expected in the near future. It may or may not change the parameters involved in determining the stock levels. With this uncertainty in mind, it is felt that work should proceed within the current guidelines, and only be adjusted if required.

Changing Technology. Several aspects of technology will influence the munition mix, and, by extension, the stock levels. Precision guided munitions, particularly those fired from traditional indirect fire platforms, such as artillery guns, howitzers, and mortars, extend the targets of these munitions beyond those they traditionally engaged. The Operational Stock Planning Process can consider these and also be used to assess the potential impact of their acquisition on stock levels. An essential tool for this would be war gaming.

Another aspect of technology is the gradual introduction of insensitive munitions. These should not affect munition performance or lethality directly, but should help diminish the risk of logistic loss, which is considered explicitly in the Operational Stock Planning Process.

Cost. Ammunition cost figures can be included in the Operational Stock Planning Process if required. It is foreseen that they would be the final stage, after the stock levels have been determined, based on the operational requirement.

Training in Theatre. Munitions consumed in theatre for training are not included explicitly in the Operational Stock Planning Process, since they are provided from a different funding envelope. They could be so considered if required. In any event, they would need to be identified for any specific operational deployment. This topic should receive some consideration, as some natures of munitions that have substitute training rounds would have operational rounds consumed in theatre for training. A not inconsiderable number of service armour-piercing fin-stabilized discarding-sabot main armament tank rounds were fired in the Gulf War by coalition forces prior to the commencement of hostilities.

Stockpile Risk. Another aspect of risk involves the ability of the stockpile to sustain operations, particularly if the planning period (30/30) or number of deployments, such as two or more consecutive 30/30 day deployments are launched. This topic is somewhat beyond the scope of this study, but it is felt to be of sufficient importance to warrant mention. This risk probably should be addressed and, if possible, quantified.

Leopard tank

DND Photo IS2003-1212a by Master Corporal Paul MacGregor

A Leopard tank belonging to The Royal Canadian Dragoons during Exercise “Resolute Warrior” at CFB Wainwright, Alberta, April 2003.

Top of Page

The Operational Stock Planning Process

The Operational Stock Planning Process proposed by the authors is an explicit decision process that incorporates professional military judgement, war gaming output, historical precedent, and an activity-based model to determine operational stockpiles. The underlying model is fundamentally different than those embedded in MASPP and ACROSS, in that it employs a blend of target-oriented and level-of-effort methodologies. The process comprises three stages: the Scenario Development Stage, the Stockpile Estimation Stage, and the Approval Stage.

  • Scenario Development Stage. In this initial stage, the operational scenarios must be defined. These will be based on the eleven Force Planning Scenarios defined by the VCDS. These scenario descriptions, however, lack sufficient detail to be directly useful within the Operational Stock Planning Process or even to drive the requisite war gaming. This stage certainly requires ‘sign off’ or approval from the Joint Intelligence (J2) and particularly the Joint Operations (J3) staff, as well as knowledge of and acceptance by the Joint Logistics (J4) staff. The effort required at this stage would not be insignificant.

  • Stockpile Estimation Stage. Given the scenario(s) defined in the first stage, this second stage will determine a scale based on a proposed model known as OS-Calc (short for Operational Stock Calculator). As mentioned above, this model is a blend of level of effort and target- oriented methodologies. It differs from MASPP in several ways.

  • Breakpoints. OS-Calc makes provision for conflict breakpoints. Whereas MASPP (without modification) requires that 100 percent of enemy assets be killed, OS-Calc assumes that hostilities will stop when a user-specified percentage of these assets has been destroyed. For instance, the Joint Munitions Effectiveness Manual suggests a stop-loss criterion of 30 percent.29 The empirical evidence presented in Helmbold’s Decision in Battle suggests that, historically, most breakpoints have occurred at a considerably lower percentage.30

  • Suppression and Speculative Fire. OS-Calc makes provision for suppression and speculative fire. While it is certainly true that a level of effort approach would include a provision for such fire implicitly, these are treated explicitly by OS-Calc. Moreover, the user would be allowed to adjust the quantity of these kinds of fire with a simple parameter adjustment.

  • Logistic and Attrition Loss. OS-Calc makes provision for logistic and attrition loss. In warfare, the ability of an army to direct its munitions on enemy assets is degraded over time as its assets are destroyed or neutralized by the enemy. Again, OS-Calc would allow a user to specify the logistic and attrition losses by adjusting a single parameter. This capability is considered by some as capability modeling.

  • Model Risk Premium. Finally, OS-Calc makes provision for the possibility that its parameter estimates are incorrect. The methodologies available for stockpiling calculations, including MASPP, ACROSS, and OS-Calc, are significant abstractions of the reality of war and combat. Hence, it is essential that OS-Calc parameters be validated by more realistic models of conflict, such as wargames, historical precedent, and professional judgement. Based on the uncertain OS-Calc parameter estimates coming out of these war games, it is suggested that operational stocks should be incremented by an amount justified by this uncertainty. This factor is termed the Model Risk Premium. By way of example, suppose that OS-Calc, excluding considerations of model risk, requires the CF to have 500 units of a particular munition in inventory. Then the final step in the OS-Calc calculation would require that this amount be incremented by the amount of the Model Risk Premium. If the Model Risk Premium is judged to be 60 units, then, ultimately, OS-Calc would recommend that the operational stock level be 500 + 60 = 560 units.

There are several good reasons for validating OS-Calc parameter input with war gaming output. Like MASPP, the essential feature of the OS-Calc model is an allocation of targets to munitions. There are factors exterior to OS-Calc that will affect these allocations. One is terrain. Another is the enhanced battlefield awareness system that the CF will incorporate over the medium term. War gaming is felt to be essential to ensuring that these munition-to-target allocation parameters are set properly. Another good reason for war gaming is to make sure that there is sufficient capability to deliver the munitions. Like MASPP, OS-Calc assumes that there are enough weapon system platforms to deliver the munitions required to destroy a given target set. The best way to check this assumption is with war gaming.

Relative to MASPP, it is considered that OS-Calc would result in significantly lower stockpile levels. While it is true that the three categories, Suppression and Speculative Fire, Logistic and Attrition Loss, and Model Risk Premium will increase the estimate of the number of rounds required relative to MASPP, this increase is far outweighed by OS-Calc’s recognition of breakpoints significantly lower than 100 percent. Initial experiments with OS-Calc suggest that it will produce stockpiles of the order of 50 percent of those calculated using MASPP.

Finally, the target-oriented model within OS-Calc is essentially the same model used by MASPP, except that the computation is done without resort to a ‘Monte Carlo’ or random-number-based simulation.

  • Approval Stage. This final stage requires the application of professional judgement and possibly war gaming and historical precedent to validate the Scale produced by the calculations in the second stage. It is worth pointing out that the Operational Stock Planning Process offers a convenient way to present decision-makers with the trade-off between risk and cost. A simple table presenting the stockpile cost for breakpoints of 30 percent, 40 percent and 50 percent might be sufficient.

Top of Page


This paper has assessed the two main candidates for stockpile calculations, MASPP and ACROSS. Both employ primarily a target-oriented methodology. While the underlying algorithm in ACROSS is not readily available, it is clear that MASPP can be improved with some simple additions including breakpoints, suppression and speculative fire, logistic and attrition loss, and a provision for model risk. These additions are offered in the Operational Stock Planning Process, an explicit decision process that incorporates a blend of target-oriented and level-of-effort methodologies, a decision process that is validated throughout with war gaming results.

The essential ingredient to the Operational Stock Planning Process is scenario definition. Without well-defined scenarios that capture the main elements of risk, the process will be of limited value. Of some importance, here, is the content of the next Defence White Paper. Care should be taken in the scenario definition stage to ensure as much as possible that the chosen scenarios are sufficiently robust to handle the likely capability required by any new defence policy.

It should also be noted that the Operational Stock Planning Process could be used to assist in planning for a specific operation. This would be especially true once the process had been run to determine overall operational stock levels.

Finally, it should be emphasized that the Operational Stock Planning Process and the underlying model are designed to calculate operational stocks for user-defined scenarios consistent with the CF Force Planning Scenarios and the 60-day sustainment requirement for the Main Contingency Force. In the event there is a relatively large conflict or sequence of conflicts, exhausting these operational stocks before these operations have been completed, it is unlikely that additional sustaining stockpiles could be produced domestically quickly enough to be of use. In view of the structure of the domestic ammunition supply industry, some thought should be given to innovative ways of dealing with this risk.

This stockpile planning process has been developed in outline to address a number of concerns identified with current tools (ACROSS and MASPP). This process should be applied to a specific but limited suite of battle-decisive munitions, to refine the methodology, including defining the operational scenarios required for analysis. Once the process has been completed, the methodology and results can be assessed to identify any modifications required.

Once the methodology has been established and accepted, including the scenarios, the remainder of the stock items can be assessed.

At this stage, the Operational Stock Planning Process should be available for use both to establish stock levels for individual munitions about to come into service, and for specific deployment scenarios.


The authors wish to acknowledge the staffs of the Directorate of Land Forces Structure (Colonel C.R.D. Davies and Major C.A. Heron), of ADM (Materiel) (Major S.H. Jennings and Major J.R.N.J. Richardson) and of the Directorate General Operational Research (Ivan Taylor) for their assistance.

CMJ Logo

Dr. William S. Andrews teaches in the Department of Chemistry and Chemical Engineering at the Royal Military College of Canada. Dr. William J. Hurley teaches in the Department of Business Administration at RMC.


  1. Defence Plan 2001, DND Ottawa, 2001 and Defence Planning Guidance 2001, DND Ottawa, 2001.
  2. Defence Planning Guidance 2001.
  3. Directive for Conduct of Review of Ammunition and Sonobuoy Requirements, 11300-1(DGFD), 29 July 1993.
  4. Ammunition Planning Working Group Zero-Based Ammunition and Sonobuoy Review Report to DMC – Executive Summary, 1993.
  5. J. Van Horn, A Validation of Canadian Forces Land Ammunition Scales, ORA Project Report PR648, DND, October 1993. See also P.A. Vincent, L.A. Willner, and J. Van Horn, An Examination of Ammunition Expenditure Rates for Low to Mid Level Operations, D Log A Project Report PR 668, DND, December 1993.
  6. Van Horn, A Validation of Canadian Forces Land Ammunition Scales.
  7. Force Development – Ammunition Planning Working Group, 24 March 1995.
  8. Force Development – Ammunition Planning Working Group, 20 June 1995.
  9. Ammunition Planning Working Group Report, 20 June 1995.
  10. Bi-Strategic Commands Stockpile Planning Guidance SPG-2001 (Bi-SC SPG-2001), NATO, 2001.
  11. ACE Resource Optimization Software System ACROSS-LEMEM, Version 3.0 Beta IV Tutorial Aid, NATO Brussels.
  12. I. Taylor, A Case Against Linear Programming for Stockpile Planning, DOR(J&L) Research Note RN-9816, DND, June 1998.
  13. Report on the Land DEM 19-22 June 2001, Prague, NATO C3 Agency, 2001. See also Fire Support Prototype Example Briefing, ACROSS DEM LEMEM, June 2001.
  14. NATO C3 Agency, Report on the Land DEM.
  15. UK Evaluation of the Marginal Analysis Stockpile Planning Program (MASPP), DERA/CDA/FRN/SPR/1862/1, 4 May 2001.
  16. R.L. Helmbold, Decision in Battle: Breakpoint Hypotheses and Engagement Termination Data, Rand, Santa Monica, June 1971.
  17. Joint Munitions Effectiveness Manual
  18. A.N. Karabulut, Validation of Air-to-Air Missile Requirements, DOR(CORP) Research Note 2001/XX, DND Ottawa, February 2002.
  19. L. Stelmate and Y. Gauthier, Air Force Stockpile Requirements – Air-to-Air Missiles, ORD Project Report 2002/04, DND Ottawa, March 2002.
  20. A.N. Karabulut and S. Jennings, An Analysis of Land Forces Ammunition Capability Using MASPP, DOR(CORP) Research Note 2002/XX, DND Ottawa, April 2002.
  21. Sonobuoy Requirements Methodology, 28 November 1995.
  22. G.L. Christopher, Air Force Operational Stocks Requirements, ORD Project Report PR-9622, DND, October 1996.
  23. Expected Versus Actual Munitions Usage Rates in Kosovo, 28 September 2000.
  24. Comments on CAS Briefing Note to VCDS – Expected Versus Actual Munitions Usage Rates in Kosovo, 24 October 2000.
  25. Y. Gauthier, Air Force Operational Stockpile Requirements – Air-to-Ground Munitions (DRAFT), ORD Project Report 2002, DND Ottawa, April 2002.
  26. Stockpile Planning: The WAYMARKER System, PowerPoint presentation.
  27. Department of Defense Instruction Number 3000.4, Capabilities-Based Munitions Requirements (CMBR) Process, US DOD, 10 August 2001.
  28. Annex J (Logistics XXI And Power Projection) to the USAREUR Theater Plan, at http://www.hqusareur.army.mil/nurevision%20files/htmlinks/
    , 20 August 2000.
  29. Joint Munitions Effectiveness Manual.
  30. See Helmbold, Decision in Battle.