Inventory-related refinery plan optimization multi-objective evaluation method

Abstract

The invention discloses an inventory-related refinery plan optimization multi-objective evaluation method. A refinery plan optimization objective is Max Z = sales amount-purchase cost-inventory cost, and the inventory cost = storage cost-asset value fluctuation + sales loss. The inventory cost model is more accurate, the part exceeding the upper limit of the inventory is depicted by using the storage cost, and the part below the lower limit of the inventory is depicted by using the sales loss. According to the scheme, crude oil price change factors are considered, crude oil is regarded as assets, asset value fluctuation is increased, and surplus inventory can bring benefits to enterprises.

Description

Technical Field

[0001] The present invention relates to the field of refinery production optimization and belongs to the field of new generation information technology. Specifically, it is a multi-objective evaluation method for refinery plan optimization related to inventory. Background Art

[0002] In refinery plan optimization, an oil refinery purchases N different types of crude oils and converts them into M different types of refined oils or other petroleum products (collectively referred to as petroleum products hereinafter) through a series of processing processes. Each petroleum product needs to be produced by mixing different types of crude oils in a specific proportion. In addition, the market demand and selling price of each petroleum product are known. This is a typical multi-objective optimization problem. The optimization objectives usually include:

[0003] The lowest procurement cost, selecting the combination of crude oils with the lowest price;

[0004] The highest sales amount, selecting the product combination with the highest profit;

[0005] A reasonable inventory level, avoiding excessive or too low inventory, reducing inventory holding costs and out-of-stock losses, etc.

[0006] Among them, according to the plan optimization, each purchased crude oil has its own upper and lower inventory limits.

[0007] In the prior art, the inventory cost model is mostly used to optimize the refinery plan, and its solution is as follows:

[0008] To unify the dimension, the inventory quantity is first converted into inventory cost. The inventory cost model is usually simplified by setting penalty values for the inventory quantity exceeding the upper or lower limit, as follows

[0009]

[0010] Among them, the over-upper-limit penalty coefficient and the over-lower-limit penalty coefficient are given by the refinery according to historical production experience. The penalty coefficient can be understood as how much storage cost is equivalent to each ton of crude oil exceeding the limit.

[0011] Therefore, the refinery plan optimization objective is:

[0012] Max Z = sales amount - procurement cost - inventory cost

[0013] Limitations of the inventory cost model:

[0014] (1) The model is too simplified: The traditional inventory cost model usually only considers the penalty values for the inventory exceeding the upper or lower limit, and these penalty values often rely on production experience and lack scientific basis. In addition, the model ignores other important cost factors, such as out-of-stock cost and overstock cost, resulting in an incomplete and inaccurate estimation of inventory cost.

[0015] (2) Static cost assumption: In actual production, the crude oil price fluctuates dynamically with the market supply and demand relationship. However, traditional models usually assume that the inventory cost is static and do not fully consider the impact of price fluctuations on the inventory cost. This static assumption may lead to a large deviation between the optimization result and the actual cost. Summary of the Invention

[0016] The present invention proposes a multi-objective evaluation method for optimizing refinery plans related to inventory to solve the problems existing in the background technology.

[0017] Technical Solution:

[0018] A multi-objective evaluation method for optimizing refinery plans related to inventory, where the optimization objective of the refinery plan is: Max Z = sales amount - procurement cost - inventory cost, and the inventory cost = storage cost - asset value fluctuation + sales loss.

[0019] Preferably, the storage cost is obtained by the following formula:

[0020]

[0021] where cost storage represents the storage cost of crude oil, represents the storage cost of each type of crude oil, N represents the number of types of crude oil purchased by the refinery, and each type of crude oil is denoted as i.

[0022] Preferably, is obtained by the following formula:

[0023]

[0024] where S i is the actual inventory of crude oil i, i = 1, 2,..., N; and are the upper and lower inventory limits of crude oil i; Unit i is the unit storage cost of crude oil i.

[0025] Preferably, the asset value fluctuation is obtained by the following formula:

[0026]

[0027] where, represents the asset value fluctuation of each type of crude oil, N represents the number of types of crude oil purchased by the refinery, and each type of crude oil is denoted as i.

[0028] Preferably, if cost asset > 0, it indicates the overall value-added income of the crude oil inventory; if cost asset < 0, it indicates the overall depreciation loss of the crude oil inventory.

[0029] Preferably, it is obtained by the following formula:

[0030]

[0031] where S i is the actual inventory of crude oil i, i = 1, 2,..., N; and are the upper and lower inventory limits of crude oil i; t buy is the crude oil purchase time, t cur is the current time; H i (t) is the unit purchase price of crude oil i at time t.

[0032] Preferably, when the crude oil price rises, i.e., H i (t cur ) > H i (t buy ), the asset value fluctuation is positive, resulting in appreciation income; when the crude oil price drops, i.e., H i (t cur ) < H i (t buy ), the asset value fluctuation is negative, resulting in depreciation loss.

[0033] Preferably, the sales loss is obtained by the following formula:

[0034]

[0035] where cost sell is the sales loss, Q i is the actual production volume of crude oil product j, D j is the demand for crude oil product j, j = 1, 2,..., M, M represents the number of types of crude oil products that the refinery needs to produce, and each product is denoted as j; P j is the unit price of crude oil product j.

[0036] Preferably, when the inventory of crude oil i is insufficient, determine the product j with the highest loss sensitivity b ij and reduce its production until the inventory is satisfied; the loss sensitivity b ij is obtained by the following formula:

[0037]

[0038] where K j represents the crude oil product cost rate, and a ij represents the formulation ratio.

[0039] Preferably, the crude oil product cost rate K j is obtained by the following formula:

[0040]

[0041] Among them, C i is the purchase unit price of crude oil i, where i = 1, 2,..., N, and P j is the unit price of crude oil product j, where j = 1, 2,..., M.

[0042] Advantages of the present invention

[0043] (1) The inventory cost model is more accurate. The part exceeding the inventory upper limit is characterized by storage cost, and the part below the inventory lower limit is characterized by sales loss.

[0044] (2) Considering the change factor of crude oil price, regarding crude oil as an asset, increasing the asset value fluctuation, and the excess inventory can also bring benefits to the enterprise. Brief description of the drawings

[0045] Figure 1 It is a schematic diagram of inventory cost analysis in the specific implementation manner. Specific implementation manner

[0046] The present invention will be further described below in conjunction with embodiments, but the protection scope of the present invention is not limited thereto:

[0047] (1) Inventory cost analysis

[0048] According to the plan optimization, each purchased crude oil has its own inventory upper and lower limits, that is, the ideal inventory range. Analyze respectively what cost impacts will be brought after the inventory exceeds the upper and lower limits, as Figure 1 shown.

[0049] 1) If it exceeds the inventory upper limit, the extra part needs to occupy the warehouse.

[0050] a. There is an inventory cost, and an amount needs to be paid for this part of the inventory every day

[0051] b. Due to the real-time fluctuation of the crude oil price, if the purchase price drops, this part of the inventory is equivalent to depreciating, causing losses; if the purchase price rises, this part of the inventory is equivalent to appreciating, which can bring certain benefits.

[0052] 2) If it is lower than the inventory lower limit, some products cannot be produced according to the demand, resulting in sales losses.

[0053] (2) Storage cost cost storage

[0054] The refinery purchases N types of crude oil, and each type of crude oil is denoted as i;

[0055] S iis the actual inventory of crude oil i, where i = 1, 2,..., N;

[0056] and are the upper and lower inventory limits of crude oil i, where i = 1, 2,..., N;

[0057] Unit i is the unit storage cost of crude oil i, where i = 1, 2,..., N;

[0058] Therefore, the storage cost of each type of crude oil is

[0059]

[0060] The storage cost of all crude oils is

[0061]

[0062] (3) Asset value fluctuation cost asset

[0063] t buy is the purchase time of crude oil, and t cur is the current time;

[0064] H i (t) is the unit purchase price of crude oil i at time t, where i = 1, 2,..., N;

[0065] Therefore, the asset value fluctuation of each type of crude oil

[0066]

[0067] Among them, when the crude oil price rises, that is, H i (t cur ) > H i (t buy ), the asset value fluctuation is positive, resulting in appreciation income; when the crude oil price falls, that is, H i (t cur ) < H i (t buy ), the asset value fluctuation is negative, resulting in depreciation loss;

[0068] The asset value fluctuation of all crude oils is

[0069]

[0070] If cost asset > 0, it indicates overall appreciation income of the crude oil inventory; if cost asset < 0, it indicates overall depreciation loss of the crude oil inventory.

[0071] (4) Sales loss cost sell

[0072] When one or more types of crude oil fall below the inventory lower limit, it will inevitably lead to the inability to produce some petroleum products as required, resulting in sales losses. However, one type of crude oil is the processing raw material for multiple products. Therefore, how to make a decision at this time, that is, which products should be produced less, can minimize the overall sales loss.

[0073] 1) Definition of basic parameters

[0074] The refinery needs to produce M types of crude oil products (such as gasoline, diesel, jet fuel, etc.), and each product is denoted as j;

[0075] D j is the demand for crude oil product j, j = 1, 2,..., M, such as the diesel demand is 800 tons and the jet fuel demand is 600 tons;

[0076] P j is the unit price of crude oil product j, j = 1, 2,..., M, such as the diesel selling price is 7000 yuan / ton and the jet fuel selling price is 6200 yuan / ton;

[0077] a ij is the formulation ratio, that is, the ratio of crude oil i consumed to produce 1 ton of product j is a ij , for example, to produce 1 ton of gasoline, 0.3 tons of light crude oil and 0.7 tons of Basra crude oil are consumed;

[0078] Q j is the actual production volume of crude oil product j

[0079] 2) Minimization of sales loss due to unmet demand

[0080] The sales loss caused by the gap between the actual production volume and the demand of crude oil product j is (D j -Q j )·P j , considering all crude oil products as a whole, define the sales loss of minimizing the unfulfilled demand:

[0081]

[0082] 3) Cost rate K of crude oil products j

[0083] For crude oil product j, according to the required crude oil volume and crude oil price for its processing, the raw material cost of the product can be calculated. The ratio of the product unit price to the raw material cost can describe the cost rate of this product. A lower cost ratio may mean a higher gross profit margin for the enterprise, with more room to cover other expenses and achieve profitability. On the contrary, a higher cost ratio may mean a lower gross profit margin for the enterprise, and more effective cost control or price increase is needed to achieve profitability.

[0084]

[0085] Among them, C i is the purchase unit price of crude oil i, where i = 1, 2,..., N;

[0086] 4) Loss sensitivity b ij

[0087] A crude oil product is processed from multiple crude oils. Therefore, based on the product cost rate, the loss sensitivity for each crude oil is further defined:

[0088]

[0089] The lower the loss sensitivity, the higher the profit rate of the product under the condition of consuming the same amount of raw materials, and it should be given priority in production when the inventory is insufficient; the higher the loss sensitivity, the lower the gross profit of the product under the condition of consuming the same amount of raw materials, and it should be given priority in reducing production when the inventory is insufficient.

[0090] 5) Sales loss

[0091] When the inventory of crude oil i is insufficient, determine the product j with the highest loss sensitivity b ij and reduce its production until the inventory is satisfied. Thus, the actual production volume Q i of the crude oil product j is obtained, so the sales loss is

[0092]

[0093] Q i is the actual production volume of the crude oil product j

[0094] (5) Optimization objective of refinery plan

[0095] Max Z = sales amount - purchase cost - inventory cost

[0096] Among them, the inventory cost is:

[0097] Inventory cost = storage cost - asset value fluctuation + sales loss

[0098] = cost storage - cost asset + cost sell

[0099] (6) Example of inventory cost

[0100] 1) Basic information

[0101] A refinery currently has inventories of 4 types of crude oil (light sand, Basra, Oman, and Sakhalin), and needs to produce 5 types of products (gasoline, diesel, jet fuel, etc.). The unit storage cost of crude oil is 78 yuan / ton / month. Table 1 shows the crude oil inventory S i , the unit price H i (t buy ) at the time of purchase, and the current unit price H i (t cur ). Table 2 shows the product demand D j and the unit price P j . Table 3 shows the product formulation ratio a ij .

[0102] Table 1 Crude Oil Inventory and Procurement Costs

[0103]

[0104] Table 2 Product Demand and Unit Price

[0105]

[0106] Table 3 Product Formulation Ratio (Unit: Tons of Crude Oil / Ton of Product)

[0107]

[0108]

[0109] 2) Cost rate K j

[0110] Calculate the cost rates of all products respectively. For example, 1 ton of gasoline needs to be processed from 0.3 tons of light sand crude oil and 0.7 tons of Basra crude oil, and the processing cost is 0.3×3800 + 0.7×3200 = 1140 + 2240 = 3380 yuan / ton. The selling unit price of gasoline is 8200 yuan / ton, so the cost rate of gasoline is 8200 / 3380 = 2.43. The cost rates of all products are shown in Table 4

[0111] Table 4 Product Cost Rates

[0112]

[0113] 3) Product loss sensitivity b ij

[0114] Calculate the loss sensitivities of all products to all crude oils respectively. For example, 1 ton of gasoline needs to be processed from 0.3 tons of light sand crude oil and 0.7 tons of Basra crude oil. Therefore, the loss sensitivity of gasoline to light sand crude oil is 0.3 / 2.43 = 0.12, and the loss sensitivity of gasoline to Basra crude oil is 0.7 / 2.43 = 0.29. The loss sensitivities of all products are shown in Table 5

[0115] Table 5 Product Loss Sensitivity

[0116]

[0117] 4) Inventory Gap

[0118] Calculate the consumption of each type of crude oil according to the product demand. For example, for light Arabian crude oil, 300 tons are required for gasoline production, 120 tons for aviation kerosene production, and 40 tons for fuel oil production, totaling 460 tons. The gasoline inventory is 500 tons, and finally 40 tons remain. If production is carried out according to the plan, the remaining crude oil is shown in Table 6. Among them, 40 tons of light Arabian crude oil remain, 20 tons of Oman crude oil remain, 20 tons of Sakhalin crude oil remain. There is a shortage of 180 tons of Basra crude oil.

[0119] Table 6 Crude Oil Inventory for Planned Demand

[0120]

[0121] 5) Actual Production

[0122] For Basra crude oil, the highest loss sensitivity for asphalt is 0.7. Therefore, it is chosen to reduce the production of asphalt. The asphalt demand is 300 tons, and the actual production is 75 tons. The Basra crude oil is just sufficient, as shown in Table 7. Among them, 40 tons of light Arabian crude oil remain, 65 tons of Oman crude oil remain, and 20 tons of Sakhalin crude oil remain.

[0123] Table 7 Crude Oil Inventory for Actual Production Volume

[0124]

[0125] 6) Sales Loss

[0126] The asphalt demand is 300 tons, and the actual production is 75 tons. Therefore, the sales loss cost sell is (300 - 75) × 3800 = 855000 yuan.

[0127] 7) Storage Cost

[0128] The unit storage cost of all crude oils is 78 yuan / ton / month. Therefore, the storage cost cost storage is (40 + 65 + 20) × 78 = 9712.5 yuan.

[0129] 8) Asset Value Fluctuation

[0130] As can be seen from Table 1, the recent crude oil price has risen slightly, which can bring a certain asset appreciation for the remaining inventory. Therefore, the asset value fluctuation cost assetIt is (3900 - 3800) × 40 + (3980 - 3900) × 65 + (3720 - 3600) × 20 = 11,600 yuan.

[0131] 9) Inventory cost

[0132] Inventory cost = storage cost - asset value fluctuation + sales loss

[0133] = 9712.5 - 11600 + 85500 = 853112.5 yuan

[0134] Compared with the traditional model, the above inventory cost can be more accurate and comprehensive when depicting the optimization goal of the refinery plan. Specifically, it is reflected in the following aspects:

[0135] Accuracy: By analyzing various influencing factors of inventory cost in detail, the estimation of inventory cost is made more accurate.

[0136] Dynamic nature: Considering the dynamic changes in crude oil prices and adding the calculation of asset value fluctuations, the model can better reflect the actual market situation.

[0137] Comprehensiveness: Considering storage cost, asset value fluctuation and sales loss comprehensively, a more comprehensive inventory cost model is provided.

[0138] These advantages enable the refinery plan optimization model proposed in this application to provide more accurate and reliable optimization results in practical applications, thus helping enterprises better manage inventory and optimize production plans.

[0139] The specific embodiments described herein are merely illustrative of the spirit of the present invention. Those skilled in the art to which the present invention pertains can make various modifications or supplements to the described specific embodiments or use similar ways to substitute, but will not deviate from the spirit of the present invention or exceed the scope defined by the appended claims.

Claims

1. A multi-objective evaluation method for optimizing refinery plans related to inventory, where the optimization objectives of the refinery plan are as follows: Max Z = Sales amount - Procurement cost - Inventory cost It is characterized in that The inventory cost = Storage cost - Asset value fluctuation + Sales loss.

2. The method according to claim 1, characterized in that, The storage cost is obtained by the following formula: Among them, cost storage represents the storage cost of crude oil, represents the storage cost of each type of crude oil, N represents the number of types of crude oil purchased by the refinery, and each type of crude oil is denoted as i.

3. The method according to claim 2, characterized in that, Obtained by the following formula: Among them, S i is the actual inventory of crude oil i, where i = 1, 2,..., N; and are the upper and lower inventory limits of crude oil i; Unit i is the unit storage cost of crude oil i.

4. The method according to claim 1, characterized in that, The asset value fluctuation is obtained by the following formula: Among them, represents the value fluctuation of each crude oil asset, N represents the number of types of crude oil purchased by the refinery, and each type of crude oil is denoted as i.

5. The method according to claim 4, wherein If cost asset > 0, it indicates the overall value-added income of crude oil inventory; if cost asset < 0, it indicates the overall depreciation loss of crude oil inventory.

6. The method according to claim 4, wherein Obtained by the following formula: Among them, S i is the actual inventory of crude oil i, where i = 1, 2,..., N; and are the upper and lower inventory limits of crude oil i; t buy is the crude oil purchase time, and t cur is the current time; H i (t) is the unit purchase price of crude oil i at time t.

7. The method according to claim 6, wherein When the crude oil price rises, i.e., H i (t cur ) > H i (t buy ), the asset value fluctuates positively, resulting in appreciation gains; when the crude oil price drops, i.e., H i (t cur ) < H i (t buy ), the asset value fluctuates negatively, resulting in depreciation losses.

8. The method according to claim 1, wherein The sales loss is obtained by the following formula: Among them, cost sell is the sales loss, Q i is the actual production volume of crude oil product j, D j is the demand for crude oil product j, j = 1, 2,..., M, where M represents the number of types of crude oil products to be produced by the refinery, and each product is denoted as j; P j is the unit price of crude oil product j.

9. The method according to claim 8, characterized in that, When the inventory of crude oil i is insufficient, determine the loss sensitivity b ij For the highest product j, reduce its production until the inventory is satisfied; loss sensitivity b ij Obtained by the following formula: Among them, K j represents the crude oil product cost rate, and a ij represents the formulation ratio.

10. The method according to claim 9, wherein Crude oil product cost rate K j Obtained by the following formula: Among them, C i is the purchase unit price of crude oil i, where i = 1, 2,..., N, and P j is the unit price of crude oil product j, where j = 1, 2,..., M.

Images