BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to the estimation of changes in the availability of resources for the creation and delivery of goods and/or services resulting from the impact of a global disruption or crisis, including, but not limited to, the circumstances created by a pandemic.
2. Background Description
Under a disruption or crisis, the availability of resources affecting the production of goods and/or services by a firm may be impacted. It is important for a firm to understand how the availability of various resources necessary for the production and delivery of goods and/or services to their customers may change due to a crisis, or disruptive event, since this will ultimately affect the firm's ability to operate profitably during, and in the aftermath of the crisis.
Examples of the impact that a crisis may have on the availability of resources (also more generally referred to as “supply”) include, but are not limited to, the following:
- Reduced availability of firm's employees;
- Reduced availability of firm's partners' or suppliers' employees;
- Reduced availability of infrastructure (potentially including, but not limited to, air, water, road, telecommunications, buildings, information technology and electricity);
- Reduced availability of logistics hubs (e.g., international shipping ports and airports);
- Reduced availability of raw materials and office supplies/equipment from suppliers; and
- Reduced availability of services procured or outsourced by the firm.
Under normal (i.e., non-crisis) conditions, a firm typically considers only a subset of the resources previously listed in its business-as-usual planning process(es). For example, a manufacturing firm may use a process called ‘material requirements planning’ (MRP) for managing its manufacturing process. In MRP, typically only raw materials or parts are considered to be a constraining resource (i.e., physical items used directly in the assembly or production of the final product(s)). The premise of MRP is that a manufacturer can predict the availability of their goods, either for distribution to retailers or delivery to customers, based simply on the availability of the necessary raw materials and parts. Therefore, conventional MRP systems take as input the availability of these raw materials and parts.
Additionally, a manufacturing firm may also use a process referred to as ‘capacity planning’ (CP) to estimate its capacity for producing goods. Typically, in this context, capacity refers to both machine capacity and labor capacity. Therefore, combining MRP with CP, under normal conditions, a manufacturer may consider only machine, labor and raw material/part availability when managing their manufacturing process. While resources such as clean water, electricity, network connectivity, telecommunications and third party logistics services may also be necessary to the manufacturer's operations, they are typically assumed to be unconstrained or otherwise taken for granted under normal conditions. Therefore, these latter such resource types are typically not considered as inputs in MRP or CP systems.
Under disruptive or crisis conditions, however, resources that are not typically considered to be critical or constraining, may become critical or constraining. Thus, the availability of such resources may significantly affect the firm's ability to meet the demand for its product(s).
SUMMARY OF THE INVENTION
A firm may be able to mitigate the potential impact of a crisis on the availability of resources by implementing one or more mitigation plans. For example, in the case of a disruption caused by a hurricane, structural damage to materials stored in warehouses could be reduced by reinforcing or otherwise protecting warehouse windows. In the case of a disruption caused by a pandemic, employees could be provided with vaccinations to reduce the probability of infection. Additionally, employees could be cross-trained so that they have overlapping skills and can ‘fill in’ for absent workers in the event of a crisis. These are all examples of potential mitigation plans.
The present invention provides a method and system for estimating the availability of resources that may affect a firm's business operations as a result of a crisis or disruption, by:
- Accounting for how a crisis may impact the availability of resources that a firm typically factors into its conventional business planning processes (e.g., in manufacturing, examples of such conventional processes may include MRP and/or CP)
- Accounting for how a crisis may impact the availability of resources that a firm typically does not factor into its conventional business planning processes
- Accounting for dependencies between the availability of resources that a firm typically factors into its conventional business planning processes and those that it does not
- Assessing how one or more mitigation strategies may impact the aforementioned effects of a crisis on resource availability
- Performing at least one of the above for one or more geographical locations and for one or more time periods, accounting for potential dependencies between locations and time periods.
BRIEF DESCRIPTION OF THE DRAWINGS
The foregoing and other objects, aspects and advantages will be better understood from the following detailed description of a preferred embodiment of the invention with reference to the drawings, in which:
FIG. 1 shows the use of a supply model according to the present invention.
FIG. 2 shows a system configured according to the present invention.
FIG. 3 shows a sample output assessing the effect of a crisis on supply according to the present invention.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS OF THE INVENTION
The present invention seeks to provide estimates of the impact of crises or otherwise disruptive events on supply by extending and adapting traditional supply estimation techniques by assessing the impact of a disruption on resources that may be assumed to be unconstrained under normal conditions, and which may affect the ability of the firm to produce its product(s) and/or which may impact the availability of resources typically accounted for in business planning under normal conditions. According to the present invention, a computer estimates supply requirements by (i) receiving as input a forecast of a firm's “baseline” supply of human resources, various forms of infrastructure and raw materials, (ii) correcting the forecast to account for the impact of a crisis, while also taking into account the potential effects of one or more mitigation policies, and (iii) providing the corrected forecast of the availability of supply of human resources, various forms of infrastructure and raw materials as output, and (iv) providing an additional output that includes a forecast of the availability one or more resources during said crisis that may not have been included in the said forecast of business resource availability under baseline conditions, whose availability is derived from one or more of the following: the availability of other resources that may have been included in the said forecast of business resource availability under baseline conditions, input parameter values. For example, the availability of resource type “site-open” (described in model details) is derived from the availability of employees and the business policy that determines when a site will be made accessible. As another example, the availability resource type “hub” (described in model details) at one location is derived from the availability of “localxport” resources and human resources in one or more other locations.
The details of the supply estimation model are described next.
Supply Model Details
The supply model takes as input the “baseline” availability of each relevant type of resource, in one or more geographical locations and over one or more time periods. This baseline corresponds to the availability of resources under normal, non-crisis conditions. Outputs produced by the present invention may include a time-profile of resource availability over the planning horizon, for each resource type and each geographical location of interest. The types of resources, or supply, that may be considered by the present invention span at least the following three categories:
- Human Resources;
- Raw Materials; and
- Infrastructure, which comprises, without limitation, sub-types such as
- “localxport”: a measure of availability of infrastructure which has been defined to include Air-travel, Water, Roads, Networks and Electricity
- “hub”: a measure of the availability of global logistical hubs
- “site-open”: a measure of the availability of a firm's facilities
- “lift”: a measure of the availability of the global air freight capacity
Details regarding the method for modeling each of the above types of resources are provided in the sections that follow.
Human Resources
Human resources, or people, may be modeled as a function of the number of employees that are working on-site, working from home, or absent, in each geographical location in each time period in the planning horizon. These numbers can be obtained, for example, from an existing epidemiological model which captures human behavioral effects. Productivity factors for employees in each of these three states may also be modeled as follows:
- Xs,l,t: Fraction of employees working at on-site at location, l, and time, t.
- α: Productivity factor for employees working at site, 0≦α≦1. A default (for example, α=0.95) may be set.
- Xh,l,t: Fraction of employees working at home, at location, l, and time, t.
- β: Productivity factor for employees working at home, 0≦β≦1. A default (for example, β=0.80) may be set.
- Xa,l,t: Fraction of employees absent at site, at location l, and time, t.
- St,lpeople: Baseline supply of people at location l.
- St,l,adjustedpeople: Adjusted supply of people at location l, at time, t, due to the crisis.
Thus, the effective availability of employees at location l in time period t is given by:
St,l,adjustedpeople=St,lpeople*(αXs,l,t+βXh,l,t).
Human resources can be further categorized by, for example, job type, skill set, industry expertise, years of experience and years of education. The same general approach for estimating the impact of a crisis on human resource availability would apply in these cases.
Raw Material from Suppliers
The availability of one or more raw materials from suppliers may be modeled in terms of a linear dependence between the availability of a supplier's workforce and the ability of the supplier to deliver raw materials.
- γ: Supplier Sensitivity. This parameter models the sensitivity of supplier's capacity to people availability. It stands for the proportional drop in supplier's capacity, per unit proportional drop in net employee availability. A default value may be set (for example, a default value of 0.5 would mean that, if net employee availability at a supplier location drops by 10%, then supplier parts capacity would drop by 5%, i.e., 5%=0.5×10%).
- κ: Supplier Buffer. This parameter models the buffer that each supplier has planned for delivery of material to the firm. It stands for the proportion of the baseline supply amount, which is the buffer planned by the supplier over and above the baseline amount, for delivery to the firm. A default value (for example, 0.10) may be set.
- St,lparts: Baseline Supply of supplier parts from location l, at time, t. St,l adjustedparts: Adjusted supply of supplier parts from location, l, at time, t, due to the crisis.
Thus the adjusted availability of supplier parts in location l in time period t due to a crisis may be given by,
St,l,adjustedparts=Min(St,lparts,St,lparts*(1+κ)*(1−(αXs,l,t+αXh,l,t))*γ).
Infrastructure-Related Supply
The dependence of a firm's operations on infrastructure-related types of supply may be modeled in terms of the following items:
1. “localxport”
This sub-type models the effect of local infrastructure in a given location. It is a measure of availability of infrastructure which has been defined to include Air-travel, Water, Roads, Networks and Electricity:
- St,llocalxport: Baseline Supply of “localxport”, in location, l, and at time, t.
- Yroad,l,t: Proportional availability of Roads, in location, l, and at time, t.
- Ywater,l,t: Proportional availability of Clean Water, in location, l, and at time, t.
- Yelec,l,t: Proportional availability of Utilities and Electricity, in location, l, and at time, t.
- Ynetwork,l,t: Proportional availability of Network Communications, in location, l, and at time, t.
- Yair,l,t: Proportional availability of Air-travel, in location, l, at time, t.
- St,l,adjustedlocalxport: Adjusted supply of “localxport”, in location, l, and at time, t.
Thus,
St,l,adjustedlocalxport=St,llocalxport*Average{Yroad,l,t,Ywater,l,t,Yelec,l,t,Ynetwork,l,t,Yair,l,t}.
2. “hub”
This infrastructure resource type models the dependency of the firm on the availability of logistics hubs (typically, these are major international airports). Any given hub may service one or more geographical regions. A set of hubs may service overlapping geographical regions. The availability of a hub depends on the availability of human resources and local infrastructure in the location of the hub:
- St,lhub: Baseline Supply of “hub”, in location, l, and at time, t.
- H(l): A set of hub locations that may service location, l.
- St,l,adjustedhub: Adjusted availability of logistics hubs in location, l, at time, t.
The adjusted availability of logistics hubs in location l and time period t may be given by,
St,l,adjustedhub=St,lhub*Average{h in (H(l)}[(Yroad,h,t+Ywater,h,t+Yelec,h,t,+Ynetwork,h,t,+Yair,h,t+(αXs,l,t+βXh,l,t))/6].
3. “site-open”
This infrastructure resource type reflects on whether a site or facility is open or closed. It takes on value of 0, or 1.0 denotes a closed site, and 1 denotes an open site. When a site is closed, then the availability of one or more resources located at, or associated with, that site may be considered unavailable. Site availability (or “site-open”) is computed as follows:
- St,lsite-open: Baseline Supply availability of facility/site, in location, l, and at time, t. St,lsite-open is assumed to be equal to 1, i.e., the site is considered open, in the baseline.
- St,l,adjustedsite-open: Adjusted availability of facility/site, in location, l, and at time, t.
- η: Site closure threshold fraction. Fraction of employee absenteeism, above which, a site will be closed. A default value may be set (for example, a default value of 0.9 would mean that a site is considered closed if absenteeism exceeds 90%).
The adjusted availability of a site in location l in time period t may be given by
St,l,adjustedsite-open=1, if η<=Xa,l,t , and
St,l,adjustedsite-open=0, otherwise.
4. “lift”
This infrastructure resource type models the availability of global air freight capacity:
- Stlift: Baseline Supply of global “lift”, at time, t. This is an estimate (or actual) of air freight capacity that is typically available under non-crisis conditions at time, t, in order to deliver on the baseline demand.
- δ: Scaling factor for reduction in global air-lift supply. A default value (for example, 80% or 0.8) may be set.
- St,adjustedlift: Adjusted supply of global air-lift.
The adjusted availability of global air freight capacity may be given by
St,adjustedlift=Stlift*δ.
The present invention is capable of assessing the effects of mitigation actions on the availability of resources. For example, employee cross-training may be implemented as a mitigation policy in a supply model according to the present invention. Such a model requires as input a set of cross-trained resource types, each of which is defined in terms of regular resource types that contribute towards the creation and composition of a cross-trained type.
EXAMPLE
A cross-trained resource type named JAVA_C++ could be defined as being composed of people drawn from regular resource types, namely JAVA and C++, which contribute towards creating the cross-trained type, JAVA_C++. The set of cross-trained resource types, along with the corresponding regular resource type definitions, is input for each supply location of interest. The cross-trained resource type, which is desired as a mitigation policy in any given location, is assumed to be created from its associated regular resource types that are present in the same location.
Other user-input policy parameters include, without limitation:
- ω: Yes/No: Whether cross-training policy is in place, or not. It takes values, 0 or 1. A default value may be set (for example, a default value of zero would indicate no cross-training in place).
- χ: Cross-training extent. This is a percentage value that stands for the extent of cross-training implementation. A value of 25% means that 25% of the relevant regular resource types get cross-trained. A default value may be set (for example, a default value of zero would indicate no cross-training had occurred).
- T: Time at which cross-trained resource availability starts.
- CRl,i,t: Cross-trained resource type, i, in location, l, at time, t.
- R(CRl,i): Set of regular resources that contribute to the composition of cross-trained resource type, i, in location, l.
The model preserves head count by ensuring that the cross-trained set of resources are created from the associated regular set of resources. In other words, for each unit increase in the size of any cross-trained resource type, there is a corresponding unit decrease in the size of some regular resource type that is associated with the cross-trained resource type.
EXAMPLE
Let there be 100 Java programmers and 200 C++ programmers in the baseline set of people supply in location Yorktown Heights. Also let there be a cross-trained resource type, Java_C++, which is associated with regular resource types Java and C++. Further, let the extent of cross-training be 50%, and let the cross-training start time index (in weeks) be T=5. For weeks 1 through 4, the following resource profile is used as the baseline set of people supply.
- Java: 100
- C++: 200
- Java_C++: 0
Starting at week 5, the following resource profile is used as the baseline set of people supply, in Yorktown Heights.
- Java: 50
- C++: 100
- Java_C++: 150
Note that the head-count is conserved across the cross-training mitigation policy. There are a total of 300 heads in either case. The mitigation results from the observation that, the cross-trained resource type, Java_C++ is able to service requests for both Java as well as C++ programming tasks.
- CRl,i,t=χ* Sum{r(R(CRl,i)}[(St,l,adjustedpeople,r|without cross-training)/Nr], if t>=T, and ω=1
- CRl,i,t=0, otherwise
- St,l,adjustedpeople,r=(St,l,adjustedpeople,r|without cross-training) *χ, if t>=T, and ω=1
- St,l,adjustedpeople,r=St,l,adjustedpeople,r|without cross-training, otherwise
where
- Nr: The number of cross-trained resource types that the regular resource type, r,
contributes towards
- St,l,adjustedpeople,r|without cross-training: The adjusted (people) availability of resource type, r, at time, t, in location, l, when there is no cross-training policy in place.
The present invention is capable of estimating the availability of resources in cases where there exist dependencies between the resources being estimated. For example, the availability of resource type “siteopen” may further depend on the availability of human resources with specific job role of “facility operations”.
In this case, additional user-input policy parameters include, without limitation:
- f: Min Facility Operations threshold. This is the minimum number of Human Resource with job role “facility operations” that must be available for site to open.
To calculate the availability of “siteopen” the availability of the facility operations resource is calculated first. Let Yt,l,adjustedfacility-operations represent the adjusted availability of human resource type “facility operations”. It is calculated according to the present invention described above for human resources. The adjusted availability of the siteopen resource for a site in location l in time period t may be given by
- St,l,adjustedsite-open=1, if η<=Xa,l,t, and Yt,l,adjustedfacility-operations>=f and
- St,l,adjustedsite-open=0, otherwise.
This example illustrates a specific example of how the invention calculates resource availability when the resource depends on another resource. Resources which do not depend on other resources are called independent resources. Resources which depend on other resources are called dependent resources. To effectively estimate availability of all the resources, the supply model is first invoked on the independent resources. The output of the supply model on the independent resources is then fed in as input to the supply model on the dependent resources.
Thus, according to the present invention, there is provided a method, a system, and a machine-readable medium with instructions for a computer or other data processing apparatus to estimate the availability of one or more business resources in the event of a crisis or other disruption, by:
- Receiving as input a forecast and/or forecast data of business resource availability under baseline conditions and a data set of parameter values representing, for example, one or a plurality of business policies, crisis severity, human factors and resource dependencies;
- Determining a corrected forecast of business resource availability to account for an impact of a disruption by taking into account change in the availability of one or more resources accounted for in conventional business planning processes due to said disruption and change in the availability of one or more resources not accounted for in conventional business planning processes due to said disruption; and
- Providing the corrected forecast of business resource availability as output.
The disruption or crisis may or may not be an epidemiological crisis. Examples of resources that may typically be accounted for in the business planning processes may or may not include raw materials, machinery and/or human resources. Examples of resources that are typically not accounted for in said business planning processes may or may not include, without limitation:
- Network connectivity and/or other utilities
- Clean water and electricity
- Roads
- Maritime port and shipping capacity
- Air travel capacity
- Air freight capacity
- Global logistics hubs
- Site/facility access and/or
- Availability of human resources belonging to a firm's suppliers and/or partners.
The correction of the forecast to account for the potential impact of a crisis or disruption may further take into account dependencies between resources accounted for in conventional business planning processes and resources that are not accounted for in conventional business planning processes. The corrected resource availability forecast may also take into account any potential changes in available resources due to the effects of one or more mitigation policies. Input may be provided for one or more geographic locations and/or one or more time periods, taking into account any dependencies between locations and time periods.
Referring now to the drawings, and more particularly to FIG. 1, there is shown a set of crisis parameters 101 being used as input by a supply calculator 110, which also receives data concerning baseline resource availability 102 and mitigation policies 103. The supply calculator 110 produces an adjusted resource availability 120, which is used as input by a resource interdependency calculator 130, which receives a capacity calculating function for determining resource availability as a function of interactions of resource types 109. Finally, the resource interdependency calculator 130 produces output in the form of crisis-impacted resource availability 190 to be used in supply chain or other business planning.
FIG. 2 shows a system configured according to the present invention. A computer 200 with a machine-readable medium 201 containing machine-readable instructions 202 for the computer 200 to receive a baseline business resource availability forecast as input and to provide a forecast corrected for crisis conditions as output. Said computer 200 receives operator instructions via a keyboard 210. The computer 200 obtains the baseline business resource availability forecast data via a network 230, either directly from databases 250a, 250b, 250c or from a server 240 obtaining the baseline business resource availability forecast data from databases 250a, 250b, 250c. The computer 200 then corrects the firm's baseline business resource availability forecast data to account for the impact of a crisis and provides a forecast corrected for crisis conditions as output either in human-readable format via a computer screen 221 or printer 225 or in machine-readable format over the network 230 as input to the server 240.
FIG. 3 shows a sample output assessing the effect of a crisis on supply according to the present invention.
While the invention has been described in terms of its preferred embodiments, those skilled in the art will recognize that the invention can be practiced with modification within the spirit and scope of the appended claims.
Patent Application
20070124190
